PART I : CHEMICAL AND SPECTROSCOPIC INVESTIGATIONS
OF TRIALKYLSILYL AND TRIALKYLSTANNYL METALLOIDS.
PART I1 : THE SYNTHESIS OF POTENTIAL INHIBITORS OF
SQUALENE SYNTHETASE
by
Sunaina Sharma
A THESIS SUBMITTED IN PARTIAL FULFILLMENT
OF REQUIREMENTS FOR THE DEGREE
OF DOCTOR OF PHILOSOPHY
in the Department of
Chemistry
at Simon Fraser University
Burnaby, BC.
O Sunaina Sharma, July 1989
Simon Fraser University
All rights reserved. This thesis may not be reproduced
in whole or part, by photocopy or other means,
without permission of the author.
APPROVAL
Name: Sunaina S harma
D e w : DOC& of Philosophy
Title of Thesis: PART I: CHEMICAL AND SPECTROSCOPIC
INVESTIGATIONS OF TRIALKYLSILYL AND
TRIALKYLSTANNYL METALLOIDS. Part 11: THE
SYNTHESIS OF POTENTIAL INHIBITORS OF
SQUALENE SYNTHETASE.
Examining Comrnmittee:
Chairman: Prof. P.W. Percival
Prof. A.C. Oehlschlager, Senior Supervisor
-- - -
Pr0f.S. Ki@tlqann, Supervisory Committee
m - -" - Pmf. V.A. Snieckus, External Examiner Department of Chemistry, Guelph-Waterloo Centre for Graduate Work in Chemistry, University of Waterloo, Waterloo.
Date Approved: December 14, 1989
PARTIAL COPYRIGHT LICENSE
I hereby g ran t t o Simon Fraser U n l v e r s l t y the r l g h t t o lend
my thes is , proJect o r extended essay ( t h e t i t l e o f which i s shown below)
t o users o f t he Simon Fraser U n i v e r s i t y L ib rary , and t o make p a r t i a l o r
s i n g l e copies on ly f o r such users o r I n response t o a request from the
l i b r a r y o f any o the r un lve rs l t y , o r o the r educatlonal I n s t i t u t i o n , on
i t s own behalf o r f o r one o f i t s users. I f u r t h e r agree t h a t permission
f o r m u l t i p l e copying o f t h i s work f o r scho la r l y purposes may be granted
by me o r t h e Dean o f Graduate Studies. I t i s understood t h a t copying
o r publication o f t h i s work f o r f l n a n c l a l ga in s h a l l not be al lowed
wi thout my w r i t t e n permission.
T i t l e o f Thes i s/Project/Extended Essay
Part of TrInvestigationsoflkylsilvl
and Trialkylstannyl Metalloids.
Part 11: The synthesis of Potential Inhibitors and Squalene Synthetase.
Author: -
(s ignature)
Sunaina Sharma
( name
December 14. 1989
(date)
ABSTRACT
Low-temperature, heteronuclear (IH, 7 ~ i , 1 3 ~ 2% and 1 1 9 ~ n ) NMR
spectroscopy was used to study the constitution of triakylsilyl [(R3Si)nCuLin-1*LiX]
and trialkylstannyl [(R3Sn)nCuLin-l*LiX; R = alkyl, n=l, 2 or 3 and X = CN or Br]
cuprates derived f h m CuCN and CuBpMe2S. These studies revealed that in
composition, the metallocuprates are similar to alkylcyanocuprates (RnCuLin-l*LiCN) in
that the sequential addition of R3SiLi, R3SnLi or RLi to CuCN yields RCu(CN)Li and
R2Cu(CN)Li2. Further addition of R3MLi (M = Si or Sn, R = alkyl) leads to the
formation of hitherto unknown (R3M)3CuLi2. In the case of alkylcyanocuprates addition
of RLi beyond a RLi:CuCN ratio of 2: 1 yields no new species but gives solutions with
hydrogen abstracting ability attributable to RLi. The foxmation of mixed cuprates
R3M(Me)nCuLin*LiCN (M = Si or Sn, n = 1 or 2) was also studied by low-temperature . NMR spectroscopy. These studies revealed that like mixed akylcyanocuprates, ligands
on copper readily exchanged in these metallocuprates.
The ability of methyl as a ligand to tenaciously bind to copper resulted in the
exclusive transfer of R3Si and R3Sn ligands in additions to cc,~unsaturated enones and
1-alkynes.
Low-temperature heteronuclear ( 2 ~ , l ~ , 2% and 119~n) NMR spectroscopy
was also employed to study the cuprous salt catalyzed addition of bimetallic
[ ( M ~ ~ S ~ A I E Q , M ~ ~ S ~ A I E Q , (Me3Sn-9-BBN*OMe)'Li+ and (Me3Si-9-BBN*OMe)-
Li+; BBN = 9-borabicyclo[2.2.1]nonane)] reagents to 1-alkynes. This wtiy suggesd
that metallometallations involve initial silyl- or stannylcupration followed by
transmetallation of a vinyl-copper bond by the electrophilic metal partner generated in
sim. Hence, metallometallations could be conducted either by addition of "R3SiCuW or
"R3SnCuW to 1 -alkynes followed by addition of E t2AlC1 or Br-9-BBN, or by
preformation of (R3Sn-9-BBN*OMe)-Li+ or (R3Si-g-BBN*OMe)'Li+ and R3SnAlEt2
or R3SiAlEt2 followed by addition of 1-alkynes and catalytic amounts of cuprous salts.
The 12-dimetallic adducts produced by metallometallation of 1-alkynes contain
well differentiated 1 ,2-dianion equivalents capable of forming new carbon-carbon
bonds. This chemistry yielded mild and efficient methods for the synthesis of stereo- and
redo-defined trisubstituted olefms.
Also described in this thesis are the total syntheses of some sulfonium ion
analogues of carbocationic intermediates presumed to be involved in the dimerization of
farnesyl pyrophosphate to squalene by squalene synthetase. These sulfonium ions are
presumptive inhibitors of this enzyme.
l l i e lijt o f tliose to whom I am ituik6ted is Ibng. Among thcm, the one o f premier
importance is not an indiv idd 6ut a uenedlk and in many respects a unique institution. At
Simon Fraser University I found encouragement to s t 4 and to rib mearch unrikr the
gudmce o f and in associati-on mi& chemists of outstanding a6iCity such as I)t. f i n Cam
Oefiljdlager. w h itkxaldli aduice was d w y avaiGa6lk to me in or out ofthe rikpartment.
I am unriergreat o6liation to I)t. %K P m r o y w h e w f i o h - W cooperation and
pilibnce eased nry task immensely.
I am also dkepGjpatefd to Dr. Bnrce N Lipsfiutz o f tfie University o f Calif&.
Santa Bar6amI who hud dway.5 6een a some o f inspiration and more & e n than one wants to
admitI of information.
I wiM 6e faicing in my duty if I umclirrii! tfiis ac~pudk&ement witbut refening to
the courteous 6eliaviour and woperation o f the committee at tfie University o f ~ r i t h h
Columbia, rrp.ciuUij IX. S.O. Chan and Mrs. M. lust& for d n g thc 131; 7 ~ i ~ J c and -
119~n m s p e c t n a .
TABLE OF CONTENTS
TITLE PAGE
APPROVAL
ABSTRACT
DEDICATION
ACKNOWLEDGEMENTS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
LIST OF ABBREVIATIONS
i
ii
iii
v
vi
vii ...
xlll
xiv
xix
PART I: CHEMICAL AND SPECTROSCOPIC INVESTIGATIONS OF
TRIALKYLSILYL AND TRIALKYLSTANNYL METALLOIDS.
I. CHEMICAL AND SPECTROSCOPIC INVESTIGATIONS OF
ORGANOCUPRATES.
I. 1 Introduction
1.2 Alkylcuprates derived from cuprous halides
I. 3 Alkylcuprates derived from cuprous cyanide
1.4 Solid statelX-ray analyses
11. CHEMICAL AND SPECTROSCOPIC INVESTIGATIONS OF TRIALKYLSILYL AND TRIALKYLSTANNYLCUPRATES.
Introduction
Research proposal
11.1 Trialkylsilylcuprates derived from cuprous cyanide
11.1.1 Silicon-29 NMR studies
11.1.2 Carbon- 13 NMR studies
11.1.3 Hydrogen- 1 NMR studies
11.1.4 Lithium-7 NMR studies
11.1.5 Chemical tests
11.1.6 Gilman tests
11.1.7 Conclusion
11.2 Trialkylsilylcuprates derived from cuprous halides
11.2.1 Silicon-29, Carbon-13 and Lithium-7 NMR studies
11.2.2 Chemical tests
II. 2.3 Conclusion
11.3 Mixed trialkylsilyl and homo- and mixed trialkylstannyl-
cuprates derived from cuprous cyanide
11.3.1 Silicon-29 and Carbon- 13 NMR studies on
mixed Trialkylsilylcuprates
11.3.1 .a Results and discussion
11.3.2 Tin- 1 19, Carbon- 13 and Hydrogen- 1 NMR studies
on Homotriallcylstannylcuprates
11.3.3 Tin- 1 19, Carbon- 13 and Hydrogen- 1 NMR studies
on mixed Trialkylstannylcuprates
11.3.4 Reactions of Silyl- and Stannylcuprates with
a$-unsaturated Ketones.
11.3.5 Conclusion
11.3.6 Find Comments
m. IN SITU GENERATION OF ACTIVATED METALLO-
CUPRATES
111.1 Copper(1) mediated metallometallation of 1-alkynes 110
Proposed work 115
Effect of method of preparation of eialkylstannyllithium 116
On the efficiency of stannylalumination and stannylboronation
Effect of catalyst on stannylalumination 118
Effect of mode of addition of reagents on stannylalurnination 120
and stannylboronation
Effect of solvent on stannylalumination and stannylboronation 121
Compatibility of polar functional groups with stannylalumination 122
and stannylboration
Reactions of 1,2-cis-dimetallo- 1 -akenes with electrophiles 123
Synthesis of trisubstituted alkenes 124
111.1.10 Synthesis of pheromone of the square-necked grain 126 beetle
111.2 Mechanistic studies
IIL2.1 Results and discussion
III.L.2 Spectroscupi~ scuciies
III.2.3 Conclusion
IV. EXPERIMENTAL
IV.l Preparation of trialkylsilylcuprates,
2% NMR sample preparation
IV. 1.1 Generation of silylcuprates deived from CuCN
IV. 1.2 Generation of silylcuprates derived from CuBraMe2S
IV. 1.3 Generation of mixed silylcuprates
IV.2 13C and 1~ NMR sample preparation
IV.2.1 Generation of silylcuprates derived from CuCN
IV.2.2 Generation of LiCl-free silylcuprates derived from CuCN
IV.2.3 Generation of silylcuprates derived from CUBPM~~S
IV.2.4 Generation of silylcuprates derived from CuBr in DMS
IV .2.5 Generation of mixed silylcuprates .
IV.3 Preparation of 34rialkylsilyl cyclohexanone
IV.4 Typical procedure for silylcupration with
"PhMe2SiCu1'
IV.5 Preparation of trialkylstannylcuprates,
119Sn NMR sample preparation
n r c i A . .&. Gzixiz5an of stannylcuprates derived from CuCN
IV.5.2 Generation of stannylcuprates derived from CuBrMe2S
IV.5.3 Generation of mixed stannylcuprates
IV.6 1~ and 1342 NMR sample preparation
IV.6.1 Generation of stannylcuprates derived from CuCN
IV.6.2 Generation of stannylcuprates derived from CuBr*Me2S
IV.6.3 Generation of mixed stannylcuprates
IV.7 Preparation of 3-trialkylstannyl cyclohexanone
IV.8 METALLOMETALLATION OF 1-ALKYNES
IV. 8.1 Preparation of trialkylstannyllithium and quenching with Me1
IV.8.2 Reaction of 1-decyne with B U ~ S ~ A ~ E Q
IV. 8.3 Reaction of 1 -nonyne with (Bu3Sn-9-BBN*OMe)-Li+
IV.8.4 General procedure for the preparation of disubstituted vinyl
stannanes
IV. 8.5 Metallometallation of functionalized allcynes
IV. 8.6 Preparation of trisubstituted alkenes
IV .8.7 Synthesis of pheromone of the square-necked grain beetle 200
V. REFERENCES AND NOTES 204
PART 11: THE SYNTHESIS OF POTENTIAL INHIBITORS OF
SQUALENE SYNTHETASE
VI. INTRODUCTION
VI. 1 .O Proposed work
VI. 1.1 Synthesis of Sulfonium ion mimic 32
VI. 1.1 a Future directions
VI. 1.2 Synthesis of Sulfonium ion mimic 33
VI. 1.2a Future directions
VI. 1.3 Synthesis of Sulfonium ion mimic 34
VI. 1.3a Future directions
VII. EXPERIMENTAL
VIII. REFERENCES
LIST OF TABLES
Table Page
I. 1 3 ~ chemical shifts of silylmetals and related species in THF 34
n. Gilrnan test on CuCN derived silylcuprates 48
111. Addition of silyl or stannylmetalloids to 1-alkynes 114
IV. Efficiency of production of R3SnLi observed upon quenching 116
with Me1
V. Addition of Bu3SnAlEt2 to 1-decyne
xiv
LIST OF FIGURES
Figure
1.
2.
3.
4.
5.
6 .
7.
8.
Nomenclature of Organocopper reagents.
7Li NMR spectra of solutions containing different ratios of 5
MeLi and MeCu (prepared from Cur) in THFEt20 at -70•‹C.
7Li NMR spectra of (a) "Me2CuLiM in Et2O at -70•‹C (b) "Me2CuLiW 6
initially prepared in Et20 to which THF was added.
1 3 ~ NMR spectra of solutions containing different ratios of
PhLi and PhCu (prepared from CuI) in DMS at -lOO•‹C.
6 ~ i NMR spectra of solutions containing different ratios of
PhLi and PhCu (prepared from CuI) in DMS at -lOO•‹C.
1I-3 NMR spectra of solutions containing different ratios of
MeLi and CuCN in THFEQO at -20•‹C.
1 3 ~ NMR spectra of solutions containing different ratios of
MeLi and CuCN in THFEQO at -70•‹C.
IH NMR spectra of (a) Me2Cu(CN)Li2 (b) n-Bu2Cu(CN)Li2
(c) CuCN + MeLi + n-BuLi (d) Me2Cu(CN)Li2 +
Page
3
n-Bu2Cu(CN)Li2.
1 3 ~ NMR spectrum of Me(n-Bu)Cu(CN)Li2 recorded in
Complexes of x-Acceptor ligands.
Structure of CuMe(PMe3)2.
Structure of weCuP(t-Bu)2Li { (3THF) )I,
Structure of Ph3CuLi2.
2 9 ~ i NMR spectra of PhMe2SiLi:CuCN in various ratios in THF
at -50•‹C.
1 3 ~ NMR spectra of PhMe2SiLi:CuCN in various ratios in THF
at -70•‹C. The solutions contain LiCl.
1 3 ~ NMR spectra of PhmSiLi:CuCN in various ratios in THF
at -70•‹C. The solutions are free of LiCl.
1 3 ~ NMR spectra of PhMe2SiLi:CuCN, 1 : 1 at -70•‹C in (a) THF
(b) THF/HMPA, 1 : 1.
IH NMR spectra of PhMe2SiLi:CuCN in various ratios in THF at 42
-85OC. .
7 ~ i NMR spectra of PhMe2SiLi:CuCN in various ratios in THF 43
at -70•‹C. The solutions are free of LiCl.
2% NMR spectra of PhMe2SiLi and CuBr in various ratios in 52
THF at -50•‹C.
l 3 c NMR spectra in THF at -20•‹C (a) PhMe2SiLi (b) PhMe2SiLi: 53
CuBr, 1: 1 (c) PhMe2SiLi:CuBr, 2:1 (d) PhMezSiLi:CuCN, 2:l.
2 9 ~ i NMR spectra in THF at -50•‹C (a) (PhMe2Si)3CuLi2 + CuBr, 55 . 1:0.5 (b) (PhMe2Si)2CuLi + CuBr, 1: 1.
1 3 ~ NMR spectra in DMS at -85OC (a) PhMe2SiLi:CuBr, 2: 1 58
(b) PhMe2SiLi:CuI, 2:1 (c) PhMe2SiLi:CuBr, 3: 1
(d) PhMe2SiLi:CuI, 3: 1.
l 3 ~ NMR spectra of (a) PhMe2SiLi:CuBr, 2: 1 in THF (b) 60
PhMe2SiLi:CuI, 2: 1 in DMS; the spectra were run at -85OC.
2 9 ~ i NMR spectra of (a) PhMe2SiLi (b) PhMe2SiCu(CN)Li 68
(c) (PhMe2Si)2Cu(CN)Li2 (d) PhMe2SiCu(CN)Li + MeLi (e)
(PhMe2Si)2Cu(CN)Li2 + Me2Cu(CN)Li2.
27. 1 3 ~ NMR spectra of (a) Me2Cu(CN)Li2 (b) PhMe2SiLi (c) 69
(PhMe2Si)2Cu(CN)Li2 (d) PhMe2SiLi + MeCu(CN)Li (e)
(PhMe2Si)2Cu(CN)Li2 + Me2Cu(CN)Li2.
28. 1 3 ~ NMR spectra of (a) 2 PhMe2SiLi + MeLi + CuCN at -70•‹C 72
(b) 2 PhMe2SiLi + MeLi + CuCN at -85OC
(c) PhMe2Si(Me)Cu(CN)Li2 at -70•‹C.
29. 1~ NMR spectra of (a) Me3SnLi (b) 1.0 Me3 SnLi + CuCN 77
(c) 1.5 Me3SnLi + CuCN (d) 2.0 Me3SnLi + CuCN (e) 2.5
Me3SnLi + CuCN (f') 3.0 Me3SnLi + CuCN (g) 4.0 Me3SnLi +
CuCN, the spectra were run at -70•‹C.
30. 1%n NMR spectra of (a) MegSnSnMeg (b) Me3SnLi 79
(c) 1.0 Me3SnLi + CuCN (a) 1.0 Me3SnLi + CuBr, the spectra
were run at -70•‹C.
31. 1 3 ~ NMR spectra of (a) Me3SnLi (b) 1.0 Me3SnLi + CuCN 80
(c) 2.0 Me3SnLi + CuCN (d) 3.0 Me3SnLi + CuCN
(e) 3.0 Me3SnLi + CuBr, the spectra were run at -70•‹C.
32. 1 3 ~ NMR spectra of (a) Me2Cu(CN)Li2 (b) 1.0 Me3SnLi + 85
CuCN (c) 2.0 Me3SnLi + CuCN (d) Me3SnCu(CN)Li + Md i
(e) (Me3Sn)2Cu(CN)Li2 + Me2Cu(CN)Li2; the spectra were run
at -70•‹C.
33. 1%n and 1~ NMR spectra of (a) MegSnLi (b) 1.0 MegSnLi + 86
MeCu(CN)Li, the spectra were run at -70•‹C.
34. l 3 ~ NMR spectra of (a) MegSn(Me)Cu(CN)Li2 (b)
Me3Sn(Me)Cu(CN)Li2 + MeLi (c) (Me3Sn)3CuLi2 +
Me3Sn(Me)Cu(CN)Li2 (d) (Me3Sn)2Cu(CN)Li2 + MeLi; the
spectra were run at -70•‹C.
35. 1 3 ~ NMR spectra of (a) BugSnH + Me2Cu(CN)Li2 90
(b) 2.0 Bu3SnH + Me2Cu(CN)Li2; the spectra were run at -70•‹C.
36. 1 3 ~ NMR spectra of (a) (PhMe2Si)3CuLi2 + cyclohexenone (b) 101
PhMe2Si(Me)Cu(CN)Li2 + cyclohexenone; the spectra were
run at -8J•‹C.
37. 1 3 ~ NMR spectra of (a) MegSnAlEt2 (b) Me3SnAlEt2 + CuCN 139
(c) Me3SnCu(CN)Li + Et2AlCI; the spectra were run at -70•‹C.
38. 2~ NMR spectrum of MegSnCu(CN)Li + I - ~ H octyne + 141
Et2Al0, the spectra were run at -70•‹C.
39. 19sn NMR spectra of (a) Me3SnCu (b) Me3SnCu +
1 - 2 ~ octyne; the spectra were run at -50•‹C.
40. 2~ NMR spectra of (a) MegSnCu + 1-ZH octyne (b) Mc3SnCu 144
+ 1 - 2 ~ octyne + Br-9-BBN; the spectra were run at -70•‹C.
LIST OF ABBREVIATIONS
CH2C12
CTAB
DIBALH
(Siam)2BH
DMF
DMS
DMSO
HMPA
LAH
LDA
MEMQ
Michler's ketone
NaH
TMEDA
TMS
p-TsC1
9-Borabicyclo[3.3. llnonane
Dichloromethane
Cetyltetramethylammonium bromide
Diisobutylaluminum hydride
Disiamylborane
N,N-Dirnethylfonnamide
Dimethylsulfide
Dimethylsulfoxide
Hexamethylphosphoramide
Lithium aluminum hydride
Lithium diisopropylamide
2-Methoxyethoxymethyl chloride
4,4'-Bis(dirnethy1arnino)benzophenone
Sodium hydride
N,N,N',N'-Tetramethylethylenediamine
Tetramethylsilane
p-Toluenesulfonyl chloride
PART I: CHEMICAL AND SPECTROSCOPIC INVESTIGATIONS OF
TRIALKYLSILYL AND TRIALKYLSTANNYL METALLOIDS
CHEMICAL AND SPECTROSCOPIC INVESTIGATIONS OF
ORGANOCUPRATES
Introduction
Advances in the development of organometallic reagents have often been
accelerated by an improved understanding of the species and mechanisms involved in
reactions of these reagents. l Nowhere is this more apparent than in the organic
chemistry of copper which has emerged as the most heralded among the transition metals
in organic synthesis. The pivotal role played by copper-complexes containing
transferable ligands is attested to by the frequent reviews2 which strive to keep organic
chemists abreast of the rapidly expanding methodological advances and applications of
copper-based reagents.
The ate complexes of copper developed in the years since Gilman's initial repod
of the formation of Me2CuLi (1, Gilman's reagent) are as varied as their chemistry.
Reactivity of these reagents can be changed by altering several parameters: the ratio of
CuX (X = Br or I) to R-M (being either stoichiometric or catalytic in C U X ) ; ~ , ~ the
gegenion involved (usually M = MgX or Li and more recently Na and ~n);5 the presence
of additives such as sulfides, phosphines or Lewis acids that solubilize, stabilize or
activate;26 and lastly, the choice of solvent (almost always ethereal).l
In addition to these "lower order" (LO) cuprates of the general formula "R2CuLi"
(2), "higher order" (HO) reagents, "R3CuLi2" (3), i.e., R2CuLi + RLi, and
"R2Cu(CN)Li2" (4), i.e., 2RLi + CuCN, have begun to vie for share of the attention
afforded to cuprates. Higher order species are differentiated from lower order reagents on
the basis of formal charge associated with the copper containing center: hence, the latter
are monoanionic, while the former are Cu(1) dianions (Figure 1).7
112 CuCN +
RCu
R2CuM
RR'CuM
RCu(CN)M
R3CuM2
R2Cu(CN)M,
RRtCu(CN)M2
Organocopper reagent
lower order homocuprates
lower order mixed cuprates
lower order cyanocuprates
higher order homocuprates
higher order cyanocuprates
higher order mixed cyanocuprates
Figure 1
ALKYLCUPRATES DERIVED FROM CUPROUS HALIDES
Although organocuprates are by far the most popular organotransition metal-
containing reagents for structural elaborations, very little is actually known about the
species involved. Until the recent spectroscopic studies by Lipshutz, et a17 on Cu(I)
halide derived alkylcuprates, utilization of "R2CuLiW (2) in reality implied nothing more
than a stoichiometric representation of the precursors from which it had been prepared. A
detailed examination of the IH and 7 ~ i NMR of solutions containing a 1 : 1 ratio of MeLi
and MeCu (halide free) led this group to conclude that in THF these reagents existed as
an equilibrium mixture of free MeLi, Me3CuzLi (5) and Me2CuLi (1).7
Thus, the combination of MeLi (0.5 equivalent) with LiI-free MeCu (1.0
equivalent, as a slurry in THF), afforded a single 7 ~ i signal (Figure 2a) attributed to
Me3Cu2Li (5) implying that any equilibrium of the type shown in Scheme 1 must
heavily favour 5 (i.e., k l>> k-1).798 As a result, and in the absence of other processes,
combination of equimolar quantities of MeLi and MeCu would be expected to lead to
either the formation of 0.5 equivalents of Me3Cu2Li and 0.5 equivalents of MeLi or 1.0
equivalent of MezCuLi (1). The 7Li NMR spectrum for the 1: 1 combination of
MeLi:MeCu at -70•‹C (Figure 2c) showed two resonances of unequal intensity.7 This
required both MeLi and Me3Cu2Li (5) as well as a third species, Me2CuLi (1) to be
present. A single chemical shift (-0.38 ppm) for both 5 and 1 was rationalized by
postulating rapid equilibration between these species leading to an averaging of signals, a
phenomenon well precedene for other organometallics (e.g., MeLilMe2Zr1,
~ e L i / M e 2 ~ ~ ) ? 9
Introduction of a further equivalent of MeLi to a final MeLi to MeCu ratio of 2: 1
did not result in the formation of Me3CuLi2 (6).7,8 Rather the proportion of free MeLi
increased. Since no new species was observed it was postulated that the relative
concentrations of 5 and 1 shifted toward the latter (Figure ~ 1 . 7
These studies suggested the equilibria shown in Scheme 1.
2 MeCu + MeLi ,, kl - Me3Cu2Li k- 1 5
MeLi 11 I_ I
Scheme 1
4 PPM -2
Figure 2 . 7 ~ i NMR spectra for solutions containing different ratios of MeLi and MeCu
(prepared from Cur) in THF/Et20 at -70•‹C (a) MeLi:MeCu, 0.5: 1 (b) MeLi:MeCu,
0.75: 1 (c) MeLi:MeCu, 1 : 1 (d) MeLi:MeCu, 2: 1, (e) MeLi.
Solutions containing MeLi and MeCu (with LiI) in THFEt20 in a molar ratio of
1: 1 exhibited only one 7 ~ i signal which was attributed to a halide-containing Me~cuLi,
1.7 However, a very fast equilibration of two or more cuprate species cannot be entirely
ruled out.
In Et20, MegCu3Li2 (7, derived from the combination of 1.66 equivalents of
MeLi and 1.0 equiv. Cur), was the major species in solution.lO Addition of 0.34
equivalent of MeLi (which increases the ratio of MeLi to MeCu to 2: 1) converted the
aggregate MegCu3Li2 (7) to a different form of "Me2CuLiW (1) since no equilibrium as
postulated in THFEt2O was obsmed.7 More remarkable was the observation that
"Me2CuLiN prepared initially in Et2O (without LiI) to which THF was added, was
spectroscopically different (Figure 3) than that formed initially in the same final
Et2OlIMF ratio (compare Figure 2c with Figure 3b).7
"Mo2CuLiW in Et20 "Me2CuLiM in Et,O t. THF
0.083 -0.434
Figure 3 . 7 ~ i NMR (a) "Me2CuLi" in Et20 at -70•‹C (b) "Me2CuLi" initially prepared
in Et20 to which THF was added; the spectra were recorded at -70•‹C.
Existence of free MeLi in the solutions of cuprates in THFEt20 was supported by
chemical reactivity studies. Side-by-side reactions were conducted on cyclohexanone (8,
Scheme 2) and methyl benzoate (9, Scheme 3), each being treated with MeLi,
"Me2CuLi" with LiI and "MgCuLi" without LiI. As expected, "Me2CuLi" with LiI
Scheme 2. Addition reaction of "Me2CuLin reagents to cyclohexanone, 8.
MeLi 76%
Me.guLi+LiIinTHFlEt@ 0%
M e (no LiI) in THF I Et@ 18%
did not react with either electrophile under these reaction conditions. The absence of LiI,
however, led to substantial amounts of products attributable to 13-addition of ~ e ~ i . 7
Scheme 3. Addition reactions of "Me2CuLin reagents to methyl benzoate, 9.
MeLi 15% 9%
Me2CuLi (no LiI) in THF I Et@ 10% 8%
M G u L i (no LiI) in Et@ 4% 1%
In support of the 7 ~ i NMR results, treatment of methyl benzoate with "Me2CuLi"
(LiI-free) in Et20 gave negligible 12-addition product while "Me2CuLi" (no LiI) in
THFEt20 yielded 8 1 of the 12-adduct (Scheme 3).7
In conclusion, it is clear that several forms of "Me2CuLiW exist (vide supra), the
relationships among these are summarized in Scheme 4. The presence of added lithium
salts and solvent may effect the nature of lower order cuprates and can be the determining
factors in the outcome of the reactions. It is also apparent that in ethereal solvents (e.g.,
Et20 or THF), R3CuLi2 (3) produced using copper halides, is in reality a mixture of
lower order cuprate "R2CuLiW (2) and free organolithium. Such preparations exhibit
reactions characteristic of free RLi.
MyCuLi (with LiI) Me3Cu2Li +
MeLi
THF / Et20 I MeLi + Me,Cu,Li - 2 M ~ C U L ~ (no LiI)
Scheme 4
7 +
MeLi
The reagent having the stoichiometry Ph2CuLi-PhLi appeared to be more reactive
than Ph2CuLi (10) in metal-halogen exchange reactions and coupling with aryl
bromides. This led House et al. to propose the existence of PhgCuLi2 (11).11
However, this reagent has not been confirmed spectroscopically in THF. In THF, ratios
of PhLi to PhCu in excess of 1:l contain free PhLi rather than forming 11 (Scheme 5),
whereas in dimethyl sulfide (DMS), 11 is an identifiable species which exists as a halide-
fkee higher order copper ~~ecies.12
PhLi / DMS PhCu + PhLi ,- Ph,CuLi + Ph3CuLi2
Scheme 5
In DMS, Ph2CuLi (in the presence of LiI) exists as two different complexes that
can be studied by NMR below -800c.12 The two sets of signals are attributed to
Ph2CuLi (10) in equilibium with LiI containing species, Ph2CuLi-LiI (12).12
At -lOO•‹C in DMS, the 1 3 ~ NMR spectrum (Figure 4a) of P ~ Z C U ~ L ~ , prepared
&om CuI and two equivalents of ~h6Li consisted of eight lines: four major (6161.9,
ipso; 143.0, ortho; 128.5, meta; 127.4, para) and four minor (163.1, 143.6, 128.1,
127.0 ppm) due to two different Ph gro~~s.12 These were assigned to Ph2CuLi-LiI
(12) and Ph2CuLi (10) respectively. Substitution of CuBr for CuI gave halide-free
Ph2CuLi (10) owing to the precipitation of LiBr from DMS. The four peaks in the 1 3 ~
spectrum of this reagent were at precisely the same positions as the peaks of the minor
Ph2CuLi species derived from CuI (Figure 4% inset, only shown for the ips0 carbon). 12
The 6Li NMR spectra of Ph2Cuti prepared from Cul and CuBr (Figure 5a) were
in harmony with the 1 3 ~ NMR results. The 6 ~ i spectrum of the reagent prepared
180 170 160 150 140 130 120
PPM
Fi y r e 4. 1 3 ~ NMR spectra for solutions containing different ratios of PhLi and PhCu
(prepared from CuI) in DMS at -lOO•‹C (a) PhLi:PhCu, 1:l (b) PhLi:PhCu, 2:l (c)
PhLi:PhCu, 3: 1; insets are for reagents prepared from CuBr.
5 4 3 2 1 0 -1 -2 - 3 ' - 4 - 5
PPM
Figure 5 . 6 ~ i NMR spectra for solutions containing different ratios of PhLi and PhCu
(prepared from CuI) in DMS at -lOO•‹C (a) PhLi:PhCu, 1:l (b) PhLi:PhCu, 2:l (c)
PhLi:PhCu, 3: 1 ; insets are for reagents prepared from CuBr.
from PhLi (2.0 equivalents) and CuI consisted of two peaks; the major one at 6 1.09 was
assigned to the LiI-complexed copper species (12) while the minor one (6 -0.57) was
attributed to the halide-he complex (10). The 6Li NMR spectrum of CuBr derived
Ph2CuLi (minus LiBr) showed a singlet at -0.57 pprn (Figure 5a, inset).lt
The l 3 ~ NMR spectrum of PhgCuLi2, prepared from 3.0 equivalents of PhLi
and either CuI or CuBr, contained seven major peaks (6 169.3, 164.1, 143.2, 142.2,
128.0, 126.4, 124.9, Figure 4b), neither of which corresponded to free PhLi, Ph2CuLi
(10) or Ph2CuLi-LiI (12). The two sets of signals coalesced to one set of four lines at - -80•‹C and were concentration dependent. The possibility of a slow exchange between
Ph3CuLizoLiI and Ph3CuLi2 was ruled out because the same set of signals was obtained
in halide-free preparations (compare Figure 4b with 4b inset). l2 These two sets of
signals were attributed to different aggngation states of Ph3CuLi2 (11). The 6Li NMR
spectrum of 11 consisted of a broad multi-humped peak spanning the region assigned to
LiI (2.3 ppm) and halide-free Ph3CuLi2 (-0.08 ppm, Figure 5b). Ph3CuLi2 derived
from CuBr exhibited a major singlet at 0.03 pprn (Figure 5b, inset).l2 The presence of a
signal for LiI at 2.3 pprn in the latter spectrum indicates that exchange between lithium
containing cuprates and free LiI is slow on the NMR time Alternatively, the
signals at 6 2.3 and 0.9 pprn may correspond to different aggregation states of 11 which
is also suggested by the NMR (vide supra). Fast interaggregate exchange would
explain the broad signals at -lOO•‹C.
While no free PhLi is detectable by NMR spectroscopic techniques in the
solutions of Ph3CuLi2 prepared from CuI (Figure 4b) substantial amounts of PhLi are
detected both in the l 3 ~ (Figure 4c) and 6 ~ i (Figure 5c) spectra of PhgCuLi2 (11) +
PhLi.
It appears that "Ph2CuLi", like "Me2CuLiU, is a more complex species than its
straightforward preparation indicates. In DMS, "Ph2CuLiW prepared with CuI consists
primarily (>70%) of the LiI-associated copper species, Ph2CuLioLiI (12), while CuBr
derived phenylcuprates in DMS are free of LiBr due to precipitation of this salt in this
solvent (Scheme 6).12 Futhermore in DMS, Ph3CuLi2 (11) is an identifiable higher
order reagent and not a mixture of PhLi and Ph2CuLi as is the case in THF.
X= I, Br, THF Ph2CuLi - Ph3CuLi2
11
Scheme 6
PhLi + PhCu.LiX X= Br, DMS
- - Ph2CuLi + LiBr
Structural determinations by spectroscopic techniques and a
heightened demand for ligand efficiency have led to the development of mixed cuprates
wherein the allcyl residues exhibit significantly different transferabilities. Thus, use of
non-transferable or residual, (RrLi) ligands permits full utilization of potentially valuable
transferable, (RtLi) ligands. Such mixed cuprates RtR&uLi (13) are generally prepared
via prior formation of R&u (14) to which the organolithium RtLi possessing the ligand
of interest is added (Scheme 7).
C- -
- 1 0
b-
Ph3CuLi2 - Ph2CuLi.LiI 11 X= I; DMS
1 2
Scheme 7
The most popular types of non-transferable ligands are the 1-alkynyl
residues. Cuprates possesing these ligands include pentynylcopper, C3H7M-Cu
(15a),13a t- butyl-ethynylcopper, t-C4HgC=C-Cu (15 b),13a (3-methyl-3-
methoxybutyny1)-copper, CH~OC(CH~)~C=C-CU (15c),13b
trimethylsilylethynylcopper, (CH3)3SiGC-Cu (15d),13c and 3-(dimethylamino)-
propynylcopper, (CH3)2NwOC-Cu (15e). 13d Each has its advantages and
disadvantages with regard to selectivity of transfer, solubility, and cost effectiveness.
Other mixed cuprates which provide alternatives to homocuprates include heteroatom
containing ligands e.g., t - ~ & I g 0 , ~ ~ a . b and thienyll'k>d groups. Hetermuprates formed
from lithium diphenylphosphide, (C@g)2PLi, and lithium dicyclohexylarnide,
(C@11)2NLi, are also being used due to their increased thermal stability.15 The most
recent and widely accepted non-transferable ligand is nibile. l6 This ligand, isoelecmmic
with acetylides, is the only member of this class of ligands which does not require
manipulation of an air sensitive copper-containing precursor prior to addition of RtLi
(Equation 1).
Equation 1
These alkylcyanocuprates not only deliver exclusively the Rt
moiety in substitution and addition reactions attributed to cuprates, but also exhibit higher
selectivity compared with dialkylcuprates derived from cuprous halides (i.e., Me2CuLi,
selectivity compared with dialkylcuprates derived from cuprous halides (i.e., Me2CuLi,
1) in these reacti0ns.l This higher selectivity has been recently attributed to the
stabilizing rr-accepting nitrile ligand. l6l
ALKYLCUPRATES DERIVED FROM CUPROUS CYANIDE
When organocuprates were derived from MeLi and CuCN in THF, no
equilibrium among MeCu(CN)Li (16), MezCu(CN)Li2 (17) and free MeLi (Scheme 8)
was evident.l7 Thus, when equimolar amounts of MeLi and CuCN were mixed in THF
at -20•‹C a IH signal attributed to 16 was obsemed (Figure 6a). Introduction of a funher
equivalent of MeLi to solutions containing MeCu(CN)Li (16) produced a signal assigned
to 17 (Figure 6d). Addition of MeLi to solutions of Me2Cu(CN)Li2 (17) result& in
formation of no new species but gave solutions exhibiting signals attributable to both free
MeLi and Me2Cu(CN)Li2 (Figure 6e).
The 1 3 ~ NMR spectra of the various ratios of MeLi to CuCN are in agreement
with the IH spectral data.17b Thus, these spectra were consistent with the sequential
formation of MeCu(CN)Li (16,l: 1, Figure 7b) and Me2Cu(CN)Li2 (17,2: 1, Figure
7c). Addition of a further equivalent of MeLi to 17 resulted in the formation of no new
species (Figure 7d). Addition of CuCN to 17 regenerated 16 (Figure 7e) suggesting that
the species derived from MeLi and CuCN (analogous to lower order cuprates, vide
supra) are also in a dynamic equilibrium. 17b
m -1 -2.5
PPM
W y r e 6. 1~ NMR spectra for solutioni containing different ratios of MeLi and CuCN
in THF/Et20 at -20•‹C (a) MeLi:CuCN, 1:l (b) MeLi:CuCN, 4 . 5 : 1 (c) MeLi:CuCN,
>IS: 1 (d) MeLi:CuCN, 2: 1, (e) MeLi:CuCN, 3: 1.
5 0 -5 -1 0 -1 5 -20 PPM
Figure 7. 1 3 ~ NMR spectra for solutions containing different ratios of MeLi and
CuCN in THFjEt20 at -70•‹C (a) MeLi (b) MeLi:CuCN, 1: 1 (c) MeLi:CuCN, 2: 1 (d)
MeLi:CuCN, 3: 1 (e) Me2Cu(CN)Li2 + CuCN.
kl MeLi + CuCN - * MeCu(CN)Li
k- I
k2 cucN MeLi
LiCN + Me3CuLi2 Me&u(CN)Li2
Scheme 8
The ability of ligands on copper to exchange as well as the formation of mixed
higher order cuprates of the general formula RRfCu(CN)Li2 (18) were also studied by
1~ NMR spechoscopy.17a These studies indicated that stoichiometry represented by
Me(n-Bu)Cu(CN)Liz (19) could be obtained by sequential addition of MeLi and n-BuLi
to CuCN (Figure 8c) or by mixture (Figure 8d) of Me2Cu(CN)Li2 (17, Figure 8a) with
n-Bu2Cu(CN)Li2 (20, Figure 8b). Most significant was the occurrence of two methyl
signals in the 1~ spectrum of 19 neither of which corresponded to 17.17a
Thus, the 1~ NMR spectrum of the preformed reagent (CuCN + MeLi + n-
BuLi) recorded at -27OC had two peaks (6 - 1.53 and - 1 S6) of similar chemical shift in
addition to a doublet of triplets centered at 6 -0.38 suggesting two different kinds of
methyl and methylene groups (Figure 8c). That the singlets at -1.53 and -1.56 ppm were
due to two different methyl groups was further substantiated by 1 3 ~ ( 1 ~ ) NMR
spectroscopy which revealed two distinct singlets at -10.99 and -1 1.10 ppm (Figure
9).17a This spectroscopic observation, taken together with the assumption that Cu(9
species are tetrahedral,l8 implied that 19 (and hence most likely 18 and 4) cannot be
monomeric unless they are square planar.17a
CuCN + MeLi + n-BuLi
C
-0.40 -1.50 PPM
Figure 8 . 1 ~ NMR spectra of (a) Me2Cu(CN)Li2 (b) n-Bu2Cu(CN)Li2 (c) CuCN + MeLi + n-BuLi (d) Me2Cu(CN)Li2 + n-Bu2Cu(CN)Li2; the spectra were recorded in
THF/Et20 at -27OC.
PPM
Figure 9. 1% NMR spectrum of Me(n-Bu)Cu(CN)Li2 recorded in THFEt20 at
-27OC.
It seems clear that copper halides do not react in THF with more than two
equivalents of an organolithium reagent to form stoichiometric cuprates beyond a
traditional copperu) monoanionic species (i.e., [(RZCU~)-]L~+, rather than
( R ~ c u I ) ~ - ~ L ~ + ) while the existence of CuCN derived higher order cuprates is far more
mmmon.17 Jt i s likely that the cyanide ligand permits the accumulation of negative
charge on copper in these complexes to formally produce copper@) dianions (i.e.,
[ (R~CU~(CN)]~- ]~M+) . The unusual stability19 of these reagents is attributed to the dtr
back-bonding between the nitrile ligand and copper as illustrated in Figure 10.
COMPLEXES OF n -ACCEPTOR L IGANDS
Figure 10
SOLID STATEJX-RAY ANALYSES
The sensitivity of most lithio.organocuprates to moisture and oxidation as well as
their thermal instability has seriously hampered efforts to form crystalline materials
suitable for X-ray analysis. Although successful preparations of mononuclear species
[(FQcul)-]M+ are rare, several neutral clusters have been reported.20 Most of these
contain bridging aryl groups including [ C U ~ L ~ ~ I [ ( C ~ H ~ C H ~ N M ~ ~ ) ~ ] ~ ~ (21) and
[(E t20)2] [ ~ i 2 ~ ~ 2 ( ~ - ~ 0 1 ) 4 ] 2 2 (22, Figure 1 1).
Perhaps the species most relevant to synthetic use of lower order cuprates is 21,
an example of a lithium cuprate containing Li and Cu in a 1: 1 ratio as a part of the central
core.21 As shown in Figure 11, this molecule has a planar tetranuclear structure in which
each aryl group asymmetrically bridges a copper lithium pair. The ortho-N,N-dimethyl-
aminomethyl residues in 21 serve as well-positioned intramolecular solvent that satisfy
the preference of lithium for tetracoordination and do not effect the basic s t r u d
, Li' 0:: cu
Figure 11
features of the cuprate. This interpretation was supported by the similar
tetranuclear structure observed for 22 in which one Et20 molecule coordinates to each Li
atom (Figure 11).~2 Solution NMR spectroscopic studies (observation of 1 ~ 7 ~ i - 1 3 ~ ) of
both 21 and 22 also point to a dimeric array involving bridging ligands to Cu and
Li.23a9b This feature is apparently intrinsic to other group 11 metal-lithium clusters
(e.g., Am2Li2, M = Ag, ~u).23c-d
These dimeric solid structures support the data on solution species of Me2CuLi
(1) and Me3Cu2Li (5) which are in turn based on solution X-ray scattering and
molecular weight measurements8 The observation of scalar coupling between 7~i -1%
in 21 and 22 imply o type interactions between ligands on copper and lithium.
Furthermore, a comparison of the solution IH NMR spectrum of 23 (6 0.15, Me
on ~ u ) 2 4 with that of Me2CuLi (1,6 -1.2), leads to the conclusion that a large upfield
shift for protons on carbon attached to copper in 1 may be due to direct interactions
between Me- and Li+, since no such interactions are possible in the lithium-free dialkyl-
cuprate (23, Figure 12, Structure as determined by X-ray analysis of
Figure 12
Recently, this has been challenged by Whangbo on the basis of
extended Hiickel calculations for dimeric MaCuLi, which predict preferential ligand
bonding to copper rather than lithium25
Placement of bulky groups on copper or sequestering of lithium by 12-crown-4
ether has led to the characterization of the monomeric complexes
[L~(THF)~][cu(c(s~M~~)~)~]~~~ and [~i(l2-crown-4)2][~u~e2]~~b respectively. A
novel heteroligand containing species MeCuP(t-Bu)2Li (24, Figure 13) is also
m0nomeric.~6c As observed for alkyl- and aryl-substituted homocuprates R2CuLi, the
copper atom of 24 exhibits a linear geometry and the Cu-C distance in 24 [1.940(6) A]
compares favorably with that found in (CuMe2)-[1.935(8) A1.2Oa However, an unusual
feature of this complex is a strong association of the lithium cation with the phosphorus
atom26d [Li-P = 2.54(1) A]. As a result, four-coordinate lithium relies on three
molecules of THF as ligands, and hence 24 is more accurately described as
[RCuP(t-Bu)~ {Li(THF)3
2 4
Figure 13
The first structural characterization of a higher order organocuprate
has recently been realized by Power et aZ.26e This compound was crystallized from a
mixture of 3.0 equivalents of PhLi and c u ~ r ~ e 2 s . 1 2 The structure can be represented
as a combination of (CuPh2)- and ( ~ u P h 3 ) 2 linked by three bridging Li+ ions
(Figurel4) i.e., [Li3Cu2Phg(SMe2)4]. The copper centers are trigonal(119.1•‹) and the
i.e., [Li3Cu2Phg(SMe2)4]. The copper centers are trigonal(119.1 O) and the average Cu-
C bond length is 1.925 for the (Ph2Cu)- ion and 2.02 .k for (Ph3cu)Z- moiety, which is
a little longer owing to the higher coordination number of the copper center. Two
different types of distokd tetrahedral coordination of the lithium centers are also
apparent. Li(1) is coordinated by two DMS and two bridging phenyl groups, whereas
both Li(2) and Li(3) are coordinated to one thioether and three phenyl rings. This
structure bears little resemblance to the previously reported clusters2o Lin+[Cu(5-
n)Ph6]' which are based solely on the association of three (Ph2Cu)- moieties with
bridging Li+ or Cu+ ions. The 1 3 ~ NMR spectrum of this molecule showed peaks
similar to those attributed by ~ e r t z l 2 to Ph3CuLi2 and Ph2CuLi.
Figure 14
In the final analysis, organocopper reagents tend to be oligomeric whereas lower
order cuprates can be either dimeric or monomeric depending upon the nature and steric
bulk of the ligands. The only higher order cuprate examined by X-ray analysis is
monomeric and in agreement with earlier cryosmpic measurements made by ~ s h b ~ 8 on
Me3CuLi2 (6).
CHEMICAL AND SPECTROSCOPIC INVESTIGATIONS OF
TRIALKYLSILYL- AND TRIALKYLSTANNYLCUPRATES
Introduction
Recently, silylcopper, [ ( R ~ s ~ ) ~ c u L ~ ~ - 1 . ~ ~ ~ 7 and stannylcopper
[(R3Sn)nC~Lin-l*LiX; R = allcyl, n = 1 or 2128 reagents have begun to receive a
~ i ~ c a n t share of the attention afforded to alkylcuprates by synthetic organic chemists.
An appreciation of their considerable synthetic utility can quickly be gained from
inspection of the numerous reactions (shown below) that these reagents effect. The range
of substitution reactions encompasses primary alkyl and acyl centem28,*9
Displacements following SN2' pathways occur in propargylic and allylic systems.
Additions to unconjugated acetylenes,30 allenedl and Michael acceptor& such as
enones, enoates or ynoates comprise a second major reaction type. As initially with
alkylcuprates, the structqres of these silyl- and stannylcopper reagents proposed to date
have been based solely on the stoichiometry of the solutions generated to achieve the
desired chemistry.
Research proposal
It is proposed to spectroscopically study the solution phenomenon associated
with the formation of silyl [(R3Si)nCuLin- l*LiX] and stannyl [(R3Sn)nCuLin-l*LiX]
cuprates and their reactions with metal salts. It is anticipated that information gained from
such studies will not only shed light on the nature of these useful synthetic reagents but
also provide insights into the differences between silyl- and stannylcuprates derived from
CuCN and CuX (where X = Br or I). The effect of added halide salts (i.e., LiBr vs. LiI)
on the composition of the metallocuprates will also.be examined.
It is also proposed to spectroscopically study the formation of mixed metallo-
cuprates (R3Si)n(R)CuLin*LiCN and (R3Sn)n(R)CuLin*LiCN derived from CuCN.
These reagents were of interest because of the possibility that they would selectively
transfer their metallo anions in preference to their alkyl anions in reactions with organic
substrates.
TRIALKYLSILYLCUPRATES DERIVED FROM CUPROUS CYANIDE
Low-temperature 2 9 ~ i , l 3 ~ , 1~ and 7 ~ i NMR spectroscopy was employed to
probe the composition of solutions generated by mixing dimethylphenylsilyllithium
(PhMe2SiLi, 25) with CuCN. Species likely to be formed in these experiments are
lower order (26 and 27) and higher order (28 and 29) silylcuprates, respectively.7
Lower Order Silylcuprates
Higher Order Silyhprates 28 (PhMe2Si)2Cu(CN)Li2
29 (PhMe2Si)3CuLi2
Silicon-29 NMR studies
Dimethylphenylsilyllithium (25) in THF was prepared by reaction of
phMe2sic133a (30) or (PhMe~~i)233b (31) and lithium metal. Preparations were
conducted at -5OC in THF and both sources of PhMe2SiLi (25) gave 2 9 ~ i signals at
-28.5 pprn (Figure 15a). Solutions of silylcuprates27b were generated by addition of 25
(prepared from 30 and Li metal) to THF solutions of copper cyanide at -50•‹C. When the
ratio of PhMe2SiLi (25) to CuCN was 1:l a major singlet at -25.5 pprn was observed in
the 2 9 ~ i spectrum (Figure 15b) which is attributed to PhMezSiCu(CN)Li (26).
Supporting evidence for this formulation comes from infrared analysis of these solutions
which shows a bqund nitrile (VCN= 21 11 cm-1)17934 but no h e LiCN or C U C N . ~ ~
As the ratio of PhMe2SiLi (25) to CuCN was raised from 1: 1 to 2: 1 (>0.5 but
~ 1 . 0 equivalent), a new peak at -24.7 pprn in the 2 9 ~ i spectrum increased in intensity at
the expense of the signal at -25.5 pprn (Pigure IX). When the ratio of PhMe2SiLi to
CuCN reached 2: 1, the major signal was that at -24.4 pprn and a minor signal at -18.8
pprn was visible (Figure 15d). The signal at -24.4 pprn is assigned to
(PhMe2Si)2Cu(CN)Li2 (28) and not (PhMe2Si)2CuLi (27) again based on the absence
0.0 -1 0.0 -20.0 -30.0 PPM
Fi y r e 15. 2 9 ~ i NMR spectra for P ~ M ~ ~ S ~ L ~ : C ~ C N in various ratios in THF at -50•‹C
(a) PhMe2SiLi (b) PhMe2SiLi:CuCN, 1 : 1 (c) PhMe2SiLi:CuCN, 1.5: 1 (d)
PhMe2SiLi:CuCNy 2: 1 (e) PhMe2SiLi:CuCNy 3: 1 (inset 15e) PhMe2SiLi:CuBry 3: 1.
of free cyanide in the solution as judged by infrared (VCN= 2123 crn-l) and l 3 ~
chemical shifts (163 ppm, vide infra). Alternative species possessing stoichiomemes of
PhMe2SiLi to Cu(1) other than 2: 1 or disproportionation of (PhMe2Si)2Cu(CN)Li2 (28)
would generate either free 25 or free CuCN neither of which is observed by 2 9 ~ i NMR
(6 -28.5) or IR (VCN= 2148 cm-11.3~b
Addition of another equivalent of PhMe2SiLi (25) to the above sample (Figure
15d) would be expected to generate free PhMe2SiLi (6 -28.5) by analogy with CuCN
based alkylcuprates. Contrary to these expectations, only a minor signal attributable to
25 was observed in this experiment (Figure 15e). Instead, in solutions containing
PhMe2SiLi (25) and CuCN in a 3:l ratio, a major signal at -18.9 ppm as well as a small
signal assigned to (PhMe2Si)2Cu(CN)Li2 (28, -24.4 ppm) were observed. The
appearance of signals for both 28 and 25 in the 29si spectrum of this solution indicates
that chemical exchange between these three species is slow on the NMR time scale.
An amactive formulation for the species exhibiting a 2% signal at -18.9 ppm is
(PhMe2Si)3CuLi2 (29). This composition is supported by the infrared spectrum of this
solution which exhibited a major absorption for free LiCN (VCN = 2085 c ~ n - l ) . ~ ~ The
formation of a species in which three PhMe2Si groups are directly associated with the
copper center requires a cyanide to be displaced.
That the formation of PhMe2SiCu(CN)Li (26), and (PhMe2Si)3CuLi2 (29) are
reversible was shown by addition of 0.5 equivalent of CuCN to the solution whose 2 9 ~ i
NMR spectrum is shown in Figure 15e. This results in the regeneration of
(PhMelSi)2Cu(CN)Li2 (28, Figure 15d). Further introduction of CuCN to this solution
results in the regeneration of the spectrum shown in Figure 15b.
Solutions prepared using 3:l molar ratios of PhMe2SiLi:CuBrMe2S exhibited a
2% spectra with signals at -19.0 ppm and -24.6 ppm (Figure 15e inset), but again no
free PhMe2SiLi (25). These signals are very close to those assigned to 29 and 28
respectively. The dynamic equilibria in which the species exhibiting signals at -25.5,
-24.4 and -18.9 ppm are participating is represented in Scheme 9.
k l PhMGiLi + CuCN - PhMe$iCu(CN)Li
LiCN + (PhMe$3i)3CuLi2 -Z (PhM@i)&u(CN)Li2
2 9 k-3 2 8
Scheme 9
The relative intensities of the 2 9 ~ i signals atmbuted to contributing species in
solutions whose spectra are shown in Figure 15 allow estimation of the positions of the
equilibria shown in Scheme 9. The equilibrium (kllk-1) between 25, CuCN and
PhMe2SiCu(CN)Li (26) lies significantly on the side of 26 (i.e., k l >>k-I). Likewise,
in solutions containing PhMe2SiLi (25) and CuCN in a 2: 1 ratio (PhMe2Si)zCu(CN)Li2
(28) is favored (k2>>k-2) to the spectroscopic exclusion of PhMe2SiCu(CN)Li (26). In
solutions comprised ot PhMe2SiLi:CuCN in the molar ratio 3: 1, (PhMe2Si)3CuLi2 (29)
predominates over (PhMe2Si)2Cu(CN)Li2 by -4: 1. This leads to the calculation of an
equilibrium constant K3. The value of Kg and the error reported were calculated by
averaging three determinations.36
It is interesting that the 2% chemical shift change obse&ed when lower order
silylcyanocuprates are converted to higher order silylcyanocuprates is opposite to what
would be expected from simple arguments based on electronegativity. Addition of
electron-rich PhMe2SiLi (25) to PhMe2SiCu(CN)Li (26) should increase the electron
density at a coordinating copper and hence at both silicons in the higher order species.
The 2 9 ~ i resonance in the latter would therefore be expected to be upfield of that in the
lower order reagents, not downfield as found. This anomolous pattern of shielding for
nuclei like carbon37 and silicon38 compared to IH has been explained by semiempirical
calculations of p-orbital "imbalancewand its contribution to the paramagnetic tern of
nuclear shielding.37b The observation that the 29% signals of higher order
silylcyanocuprates are downfield to those of lower order silylcyanocuprates has a parallel
in l 3 c NMR spectroscopy. Deshielding of the carbonyl carbon of transition metal
carbonyls increases with increased metal-to-carbonyl x back-donation. This is evidenced
by an inverse linear relationship between metal-carbonyl IR stretching constants and 1 3 ~
chemical shifts of such ~arbon~ls.39 Incnased metal-to-carbonyl x back-donation has
been attributed to a decrease in the magnitude of the separation between the ground state
and the lowest lying excited states of these molecules.39 The observation that the 2 9 ~ i
chemical shifts change in the same manner as the 1 3 ~ chemical shifts of metal-bonded
carbonyls, suggests that there is a significant x bonding in both the M-CO and the Cu-Si
bonds.
Carbon-13 NMR Studies
The chemical shifts of the 13c resonances for the silylcuprates and their
precursors are shown in Table I. Specific peak assignments were made with the aid of
proton decoupled as well as proton coupIed spectra. The major trends apparent in Table I
are that the formation of silylanions from neutral silyl derivatives caused both the ips0
and methyl carbons to be strongly deshielded and the para carbons to be shielded with
the maximum effect observed in the case of PhMe2SiLi. This effect is expected by
extension of the Cu-Si x bonding arguments advanced above to Si-C bonds. Such a
bonding would be predicted to be more significant for Si-Csp2 than for Si-Csp3. As
expected the ipso carbons of PhMe2SiLi experience larger deshielding than the methyl
carbons compared to ~ h m s i c 1 . m
Similar reasoning can be used to explain the chemical shift changes observed for
the 1 3 ~ resonance of the ips0 carbon upon the addition of electron releasing ligand
(PhMe2SiLi) to PhMe2SiCu(CN)Li. Thus, as one progresses in the series
PhMe2SiCu(CN)Li + (PhMe2Si)2Cu(CN)Li2 + (PhMe2Si)3CuLi2 + PhMe2SiLi the
ips0 carbon is progressively deshielded (152.0 + 158.6 + 164.3 -+ 166.3 ppm, Figure
16).
Table I. k Chemical shifts of silyl metals and related species in THF. All the spectra except for PhMgiCl were recorded at -70•‹C and are referenced to THF.
Ips0 Ortho Meta Para Me
The aforementioned solutions of silylcuprates~~b were generated by addition of
PhMe2SiLi (prepared from PhMe2SiC1 and Li meta1?3 Figure 16a) to THF solutions of
CuCN.2LiCl at -50'~. The 1 3 ~ NMR spectrum of PhMe$Xu(CN)Li (26) at -70•‹c,
exhibits a complex pattern consisting of a combination of sharp and broad signals in the
methyl as well as the phenyl region (Figure 16b). The sharp peaks are assigned to
(PhMe2Si)2 which is formed as a result of decomposition of 26. The broad peaks in the
1 3 ~ NMR spectrum of 26 can be attributed to either slow equilibration between species
of different silicon to copper stoichiometry or different aggregation states of 26.
The possibility that signal broadening observed in the l 3 ~ NMR of 26 was due
to slow equilibration of species possessing different silicon to copper ratios was ruled
out by parallel on solutions in which the silicon to copper ratios was systematically
varied from 1: 1 to 4: 1. These solutions exhibited clearly observable and differentiable
IH, l 3 c and 2 9 ~ i chemical shifts (vide supra).
Alternatively, the 1 3 ~ line broadening observed in the spectrum of 26 could be
attributed to slow exchange between LiCl associated and LiCl-free forms of this cuprate
was also considered. The 1 3 ~ NMR spectrum of 26 which were generated free of LiCl
was identical to the one which contained this salt (compare Figure 16b with Figure 17b).
Thus, this origin of line broadening is also ruled out.
It is also possible that the line broadening in the 1 3 ~ NMR spectrum of 26 is
due to slow exchange between PhMezSiCu(CN)Li and "PhMe2SiCu". Solutions of
PhMqSiLi + CuCN "PhMqSiCu" + LiCN
Scheme 10
PPM
Figure 16. 1 3 ~ NMR spectra for PhMe2SiLi:CuCN in various ratios in THF at -70•‹C
(a) PhMe2SiLi (b) PhMe2SiLi:CuCN, 1: 1 (c) PhMe2SiLi:CuCN, 2: 1 (d)
PhMe2SiLi:CuCN, 3:l (e) PhMe2SiLi:CuCN, 4: 1; all solutions contain LiCI.
I 1 I 170 154 145 1 - u
115 13 1.5 PPM
Figure 17. 1 3 ~ NMR spectra for PhMe2SiLi:CuCN in various ratios in THF at -70•‹C
(a) PhMe2SiLi (b) PhMe2SiLi:CuCN, 1: 1 (c) PhMe2SiLi:CuCN, 2: 1 (d)
PhMe2SiLi:CuCN, 3: 1; the solutions are free of LiC1.
PhMe2SiLi (25) and CuBroMe2S which are expected to form "PhMe2SiCu" exhibited
different 2 9 ~ i chemical shifts than solutions of 26 (vide infro). This observation coupled
with the absence of free LiCN as indicated by IR analysis further rules out the possibility
of "PhMe2SiCu" in these solutions (Scheme 10).
The last possibility considered was that the l 3 ~ signal broadening was due to
slow exchange between different aggregation states of 26 in TI-IF. This phenomena has
precedent in the alkylcuprates. The latter were found to exhibit broad 1 3 ~ and ~ H N M R
signals in ethereal solvents but sharp signals in strongly coordinating solvents such as
N,N,N',N1-tetramethylethylenediarnine (TMEDA) .~~ The observation that addition of a
coordinating solvent like hexamethylphosphoramide (HMPA) results in
Ph Me,SiLi : CuCN l3 c 111 - 70%
THF 6.6
I I I I 4- 110 135 130 125 120 115 10 6 2
PPM
HMPA I THF
I 1 1 I I 4- 140 135 130 125 120 115 10 6 2
PPM
Figure 18. l 3 ~ NMR spectra of PhMe2SiCu(CN)Li in (a) THF (b) THF/HMPA, 1 : 1.
considerable sharpening of the NMR signals (Figure 18) of 26 is consistent with
the existence of this reagent in a less aggregated state in HMPA compared to ethereal
solvents.
Warming solutions of 26 to - 2 0 ' ~ resulted in the rapid formation of (PhMe2Sib
due to decomposition of this reagent at and above this temperahue. When l 3 c NMR
spectra of 26 were recorded at this temperature no alteration in the l 3c line shapes of
this cuprate were observed suggesting the rate processes giving rise to the broadening
were too slow. Spectra below - 7 0 ' ~ could not be obtained due to appreciable
precipitation of the reagent at this temperature. This precluded determination of exchange
rates for species involved in 1 3 ~ signal broadening.
The NMR spectrum of (PhMe2Si)2Cu(CN)Li2 (28) consists of six sharp
lines: four in the phenyl region (6 158.6, ipso; 134.7, ortho; 126.8, meta; 124.9, para),
one due to the nitrile carbon (156.6 ppm) and a methyl signal at 5.4 ppm (Figure 16c).
As observed earlier, substituting PliMe2SiLi, free of LiC1, for LiC1-containing . PhMe2SiLi did not change the 1 3 ~ NMR chemical shifts (compare Figures 16c and
17c) implying that the presence or absence of halide has no noticeable effect on the
solution composition of higher order silylcyanocuprates. The presence of a single species
in the l 3 c NMR spectrum is in agreement with the 2 9 ~ i spectrum (vide supra) which
showed a single peak for this reagent (Figure 15d). A single species is similarly indicated
by the 7 ~ i NMR specmun (vide i@a) of 28 which exhibits a sharp singlet at -0.96 ppm.
In ageement with the 29% NMR analysis, the l 3 ~ NMR spectrum of
(PhMe2Si)3CuLi2 showed signals corresponding to both 28 and 29 in a -4: 1 ratio
(Figure 16d). Halide free (PhMe2Si)3CuLi2 prepared from 25 gives i k ir; &c adlic
principal lines, but they are sharper (Figure 17d). The 7 ~ i spectrum of 29 shows a broad
hump centering at 6 -0.74. This is attributed to slow intermolecular exchange of lithium
among various lithium containing species.79,12 Similar signal broadening is observed
in both 1 3 ~ and 7 ~ i NMR spectra of Ph3CuLi2 in DMS even at -lOO•‹C (vide sypra).12
Introduction of a further equivalent of 25 did not result in the formation of any
new species but rather showed signals corresponding to free PhMe2SiLi (Figure Me).
As pointed out earlier, sequential addition of PhMe2SiLi to highly aggregated 26
in THF results in formation of (PhMe2Si)2Cu(CN)Li2 (28) and (PhMe2Si)3CuLi2 (29)
which are apparently less aggregated as judged from their sharp NMR signals. Whether
28 is monomeric or dimeric is not clear. If it is monomeric there is the possibility that the
complexes could be solvated according to the equilibrium in Equation 2. With
CuL2L' (solvent) 4 * CuL2L' + solvent
Equation 2
coordinated solvent the likely geometries about Cu+ are tetrahedral or square
planar. In the uncoordinated form a trigonal array around Cu+ is likely. Of further
interest is the location of the two lithium ions in the copper cluster. The placement of two
lithium ions in a tetrahedral array could convey an element of asymmetry to the complex
and yet no such asymmetry was noted in the present studies.
A dimeric species could either be cyclic or acyclic. The cyclic arrangement must
contend with planarity requirements of two nitrile ligands. An acyclic dimer containing
nitrile should permit further associations via this ligand. The difficulty in formulating
structures of dimeric cyanocuprates is reflected in the fact that no structures have been
proposed to date for the allcylcyanocuprates.
Hydrogen-1 NMR Studies
The ease with which we were able to study the higher order silylcuprates derived
from CuCN, by 2 9 ~ i and 1 3 ~ NMR spectroscopic techniques, encouraged us to study
the formation of these species using 1~ and 7 ~ i NMR spectroscopy. The 1~ nuclear
magnetic resonance spectra were recorded at -85OC at 300 MHz in the region between 1.0
pprn and -2.0 pprn with the specific goal of observing the resonances due to the methyl
hydrogens (since the chemical shifts of the aromatic protons were too close to be of any
significance). Solutions of PhMe2SiLi (25) exhibited a singlet at 0.10 pprn (Figure 19a)
whereas solutions containing 25:CuCN in a 1: 1 ratio showed a broad signal at 0.22 pprn
which we attribute to aggregated PhMe2SiCu(CN)Li (26, Figure 19b). As the ratio of
PhMe2SiLi (25) to CuCN was increased from 1: 1 to 2: 1, a new signal at 0.02 pprn
became visible (Figure 19c). When the ratio was precisely 2: 1, the major signal observed
was at 0.02 ppm; this signal was assigned to (PhMe2Si)2Cu(CN)Li2 (28, Figure 19d).
Solutions containing 25:CuCN in 2: 1 molar ratio to which one or more equivalent of
PhMqSiLi had been added revealed only one major 1~ signal with a chemical shift
close to that assigned to 28 (Figure 19e). We assume from the 2% and l 3 ~ NMR
studies (vide supra) that the chemical shifts of the methyl hydrogens attached to silicon
for the presumed species 28 and (PhMe2Si)3CuLi2 (29) are very similar.
Lithiurn-7 NMR Studies
Dimethylphenylsilyllithium (25, free of LiCi), was prepared i h r 1 iPnMe2Si)2
(31) and lithium metal.33 Initial 7 ~ i NMR studies were conducted at -70•‹C on THF
suspensions composed of equimolar amounts of PhMe2SiLi (25,8 0.29, Figure 20a)
and CuCN. These solutions exhibited a major 7 ~ i signa141b at -0.25 pprn which was
PPM
Figure 19. IH NMR spectra for PhMe2SiLi:CuCN in various ratios in THF at -85OC
(a) PhMe2SiLi (b) PhMe2SiLi:CuCN, 1 : 1 (c) PhMe2SiLi:CuCN, 1.5: 1 (d)
PhMe2SiLi:CuCN, 2: 1 (e) PhMe2SiLi:CuCN, 3: 1; all solutions contain LiCl.
PPM
Figure 2 0 . 7 ~ i NMR spectra for PhMe2SiLi:CuCN in various ratios in THF at -70•‹C
(a) PhMe2SiLi (halide free) (b) PhMe2SiLi:CuCN, 1:l (c) PhMe2SiLi:CuCN, 2: 1 (d)
PhMe2SiLi:CuCN, 3: 1; insets are for reagents with LiCl.
assigned to PhMe2SiCu(CN)Li (26) along with a peak in the region for 25 (6 0.38,
Figure 20b) and a weak broad peak at -0.84 ppm. As expected, in the absence of LiCl
the reaction mixture was heterogeneous due to the decreased solublity of CuCN.
Addition of LiCl resulted in the disappearance of the signal due to PhMe2SiLi (25) and
in the appearance of a single peak at -0.38 ppm (Figure 20b, inset). These observations
indicate that in the absence of LiC1, exchange between 25 and 26 is slow on the NMR
time scale whereas, in the presence of LiCl exchange of 7 ~ i between various species is
rapid?$ In agreement with these interpretations, a positive Gilman t e d 2 was obtained
for the LiC1-free preparations of 26 and a negative test after LiCl was added.
Sequential addition of one equivalent of LiCl-free PhMe2SiLi (25) to
PhMe2SiCu(CN)Li (26) resulted in the appearance of a singlet at -0.96 ppm which we
attribute to (PhMe2Si)2Cu(CN)Li2 (28, Figure 20c). The presence of a single peak in
the 7 ~ i spectrum of 28 is in harmony with the 2 9 ~ i , 1% and lH NMR studies (vide
supra) indicating a symmetrical environment around Li+ in the copper spe~ies.~3 The
presence of rapidly exchanging species cannot be entirely ruled out.
Further introduction of halide-free 25 led to a broad signal centered at 6 -0.74
(Figure 204. This was assigned to (PhMe2Si)3CuLi2 (29). The broad nature of this
signal suggests averaging due to rapid chemical exchange between 28,29 and free
~ i ~ ~ . 7 , 9 , 1 2
The observation of averaged signals in the cases of 1~ and 7 ~ i spectra but not in
the case of 2 9 ~ i is a function of the frequency of the measurements and the chemical shift
differences in Hz of the exchanging ~~ecies.44 Thus for 2 9 ~ i , the chemical shift
differences for 26,28 and 29 are at least 1.0 ppm and the measurement is carried out at
79.5 MHz. If the equilibration between the species is faster than 79 Hz, an averaging of
signals will be observed in the 2 9 ~ i spectrum. For 1~ assuming a separation of the
methyl lH signals of 26,28 and 29 of -0.1 ppm and a measurement frequency of 300
MHz the averaged signals observed require exchange to be faster than 30 HZ. Thus, at
-50•‹C the species equilibrate at a rate of 30 to 79 Hz.
Some uncertainty in the measurement of equilibria from 2 9 ~ i signal intensities
comes from the negative gyromagnetic ratio36b of 29si. Under the conditions of broad-
band 1~ decoupling the 29si resonance will suffer a negative nuclear Overhauser effect
if the 2 9 ~ i nucleus is in close proximity ( O A) to a 1~ nucleus. To surpress the negative
nOe of 2 9 ~ i the decoupler was m e d on during the acquisition and off during the
relaxation delay. Another source of uncertainty in calculations of equilibrium constants
could arise from the differential Tl's of the equilibrating species. Since the proposed
species are equilibrating with a rate constant between 30 and 79 Hz which is -lox faster
than the acquisition time (0.2 s, vide inpa) for 29si NMR experiments, the effective Ti's
of the equilibrating species would be equivalent under these conditions.
Chemical tests
Perhaps the single most intriguing question which arises fiom 2 9 ~ i NMR studies
relates to the composition of solutions attributed to 29. If 29 does not have the
postulated structure but rather is composed of free 25 in rapid equilibrium with
silylcyanocuprates containing a Si:Cu ratio of <3: 1, then thepee PhMe2SiLi would be
expected to compete in reactions with the silylcyanocuprates. To test this, side-by-side
reactions were conducted on two substrates, cyclohex-Zen- 1-one (32) and 1-octyne
(35a), each was treated with 25,26,28,29 and 29 + 25. It was found that solutions
of both 28 and 29 deliver a PhMe2Si group exclusively via 1,4-addition to cyclohex-2-
en-1-one (32).32 Solutions composed of 29,25 and CuCN (molar ratio 1:l:o. 1) also
added to 32 in a l,4-manner. These results show that "~h~e2S iCu" reagents behave in
Scheme 11. Addition reaction of "PhMe$iCu" reagents to cyclohex-2-en-1-one, 32.
a fashion analogous to alkyl cuprates45 in delivery of their anionic ligands in a 1,4-
fashion even when copper is present only in catalytic concentrations. Thus 1,4-addition
of silylanions bound to copper must be faster than 1,2-addition of the unbound species to
give 34. The observation that solutions of 29 + 25 gave 33 in high regioisomeric purity
suggest that under catalytic conditions the extra PhMe2SiLi serves to rapidly convert any
lower order cuprate formed as a result of silyl transfer back to a higher order cuprate.
Next we studied the addition of silylcuprates 25,2628 and 29 to 1-octyne
(35a). In agreement with existing reports,30 we found that 26 added to 35a to yield a
mixture of 36 and 37 in a 3:2 ratio, whereas addition reactions of
(PhMe2Si)2Cu(CN)Li2 (28) and (PhMezSi)3CuLi2 (29) gave 36 in -90% isolated
yields. Under similar conditions PhMe2SiLi (25) abstracted the acetylenic hydrogen of
Scheme 12. Silylcupration of 1-allcynes.
% Composition 35b 3 6 37 %yield
PhMgiLi 100 - - 72
PhMe$iCu(CN)Li - 60 40 62
PhMe$iCu(CN)Li/HMPA - >98 <2 86
(PhM@i)$u(CN)Li2 - >98 <2 92
(Pl~Me#i)~CuLi~ - >98 <2 90
35a after 2 ~ 2 0 quench to give 35b as judged by GC-MS (70% incorporation of 2~ in
1-octyne). No addition products were observed (capillary g.c. analysis). In support of
l 3 ~ NMR results, treatnient of 1-octyne (35a) with PhMc2SiCu(CN)Li (26) in
THF/HMPA (1:l) gave exclusively 36 in >85% yield. That free PhMe2SiLi is not
present in the solutions of silylcuprates was confumed by the absence of incorporation of
*H in the vinyl products when quenched with 2 ~ 2 0 .
Gilman Tests
Support for our interpretation of the %i, 1 3 ~ 7 ~ i and IH NMR data and the
propensity of solutions of 25 and CuCN studied (1:l + 4:l) to undergo 1,4-addition to
32 and regiospecific additions to 35a was obtained from Gilman tests on THF solutions
containing PhMezSiLi (25) and C U C N . ~ ~ As originally reported by Gilman, addition of
a solution containing 4,4'-Bis(dimethy1amino)benzophenone (Michler's ketone) to an
organocuprate at room temperature followed by quenching with Hz0 and then
introduction of 12MOAc gives an intense greenish-blue coloration if RLi is present. A
Table IL Gilman test on CuCN-derived silylcuprates
I Reagents Results
25 PhMe2SiLi Positive
26 PhMe2SiCu(CN)Li negative
28 (PhMe$ i)&u(CN)Li2 negative
29 (PhMe$i)3CuLi2 negative
(PhMe$i)3CuLi2 + 25 positive
positive Gilman test was obtained for 25 in THF while a negative test was obtained for
all solutions of PhMe2SiLi containing CuCN including those where these reagents are
present in a 3:l ratio. A slight green coloration was observed for solutions containing a
4: 1 ratio of PhMe2SiLi:CuCN indicating the presence of small amounts of PhMe2SiLi
(25, Table II).
Conclusion
Comparison of the present silylcyanocuprate system with that of the
methylcyanocuprate system studied by Lipshua et a1.17 reveals several interesting
features. In both of these systems, when the ratio of RLi (R = PhMe2Si or Me) to CuCN
is unity, a nitrile containing monoanionic cuprate is formed. In neither system does this
species appear to be in eipilibrium with other species or free RLi. As the proportion of
anion is increased from an RLi:Cu ratio of 1:l to 2:l a new species, containing a bound
nitrile, is formed in both cases. In the case of Me2Cu(CN)Li2 (17) association of alkyl
residues with copper beyond this stoichiometry does not occur and further addition of
MeLi gives solutions containing free MeLi. In the case of (PhMe2Si)2Cu(CN)Li2 (28)
addition of further silyllithium gives solutions which contain negligible amounts of
PhMe2SiLi and whose 2 9 ~ i , l 3 ~ and 1~ spectra support the association of three silyl
residues with the copper accompanied by displacement of the cyanide ligand Chemical
studies on two different substrate types, as well as Gilman tests for the presence of fYee
RLi, are fully consistent with the spectroscopic data.
TRIALKYLSILYLCUPRATES DERIVED FROM CUPROUS HALIDES
In the previous section it has been demonstrated that higher order silylcuprates,
(R3Si)2Cu(CN)Li2 (28) -retain the CN moiety. This lack of metal-metal exchange
between copper and lithium (i.e., R3SiLi + CuCN ++ "R3SiCuW + LiCN) was attributed
to the likelihood of dn backbonding between the copper and nitrile ligand. l 9 More
remarkable was the observation that the addition of three equivalents of PhMe2SiLi (25)
to CuCN resulted in the unprecedented formation of (PhMe2Si)3CuLi2 (29) rather than
(PhMe2Si)2Cu(CN)Li2 (28) and free PhMe2SiLi (Scheme 9) as was the case for
alkylcuprates derived fiom CuCN (Scheme 8).17 This unusual stability of 29 was
attributed to significant n bonding between Cu and Si (vide supra). Whether such an
equilibrium exists in the case of copper halide derived silylcuprates, however, remained
to be elucidated and this was next undertaken.
Silicon-29, Carbon-13 and Lithium-7 NMR studies
Low-temperature 2%i, 1 3 ~ and 7 ~ i nuclear magnetic resonance spectroscopy
was employed to probe the composition of solutions generated by mixing PhMe2SiLi
(25) with CuX (X = Br or I). Species likely to be formed in these experiments are
silylcopper reagent (38), halide-free (27) and halide associated (39) lower and higher
order silylcuprates (29), respectively?
Silylcopper reagent 38 PhMe2SiCu
Lower Order Silylcuprates 27 (PhMe2Si)2CuLi
39 (PhMe2Si)2CuLi*LiX
Higher Order Silylcuprates 29 (PhMe2Si)3CuLi2
Reaction of PhMe2SiC1(30) with lithium metal in THF gave LiC1-containing
solutions of 25.33 Preparations were conducted at -5OC in THF and 25 gave a 2 9 ~ i
signal at -28.5 pprn (Figure 21a). Solutions of silylcuprates (27) were generated by
addition of THF solutions of 25 to CuBr-Me2S at -50•‹C. The combination of equimolar
amounts of 25 and CuBrMezS yielded a suspension in which most of the 2 9 ~ i signal
was lost (Figure 21b) presumably because the solid contained most of the silicon
containing species. The resulting 2 9 ~ i spectrum contained several signals between -10
and -25 pprn which were attributed to polymeric "PhMe2SiCuM (38). In agreement with
the observations made with 2 9 ~ i NMR, most l 3 ~ signal in the corresponding 1 3 ~ NMR
spectrum was also lost in solutions containing equimolar amounts of 25 (Figure 22a)
and CuBrMe2S. The signals that were visible were due to (PhMe2Si)2 (31, Figure
22b, identity of 31 was confirmed by recording the l 3 ~ NMR of the solution containing
both 31 and 38) formed as a result of thermal decomposition of 38.
The combination of 25 and CuBrMe2S in a 1.5: 1 molar ratio gave a nearly
homogeneous solution exhibiting a 2% signal at -24.4 pprn attributed to
(PhMe2Si)2CuLi (27, Figure 21c). The assignment of this signal to 27 was facilitated
by the observation that at a 2: 1 ratio of 25 to CuBrgMe2S the signal at -24.4 pprn
remained intense while a minor signal at -18.9 pprn appeared (Figure 21d). A signal at
-24.4 pprn has previously been observed for (PhMe2Si)2Cu(CN)Li2 (28) produced by
the reaction of two equivalents of 25 and CuCN. That (PhMe2Si)2Cu(CN)Li2 (28) and
(PhMe2Si)~CuLi (27) are, however, different species was confirmed by 1 3 ~ NMR
spectral analysis which revealed different chemical shifts for the silyl methyl and phenyl
signals in these two species (compare Figure 22c with Figure 22d). The 1 3 ~ NMR
spectrum of (PhMe2Si)2CuLi (27) consists of four resonances in the phenyl region (6
162.0, ipso; 134.7, ortho; 126.3, meta; 124.9, para) one due to DMS (17.8 ppm) and a
methyl signal at 6.04 pprn (Figure 22c) whereas (PhMe2Si)2Cu(CN)Li2 (28) exhibited
0.0 -1 0.0 -20.0 -30.0 PPM I
Figure 21. 2 9 ~ i NMR specaa for PhMe2SiLi and CuBr in various ratios in THF 2:
-50•‹C (a) PhMe2SiLi (b) PhMe2SiLi:CuBr, 1 : 1 (c) PhMe2SiLi:CuBr, 1.5: 1 (d)
PhMe2SiLi:CuBr, 2: 1 (e) PhMeaSiLi:CuBr, 3: 1 (inset 21e) PhMe2SiLi:CuCN, 3: 1 (f)
PhMe2SiLi:CuBr, 4: 1.
PPM
Figure 22. 1 3 c NMR spectra in THF at -20•‹C (a) PhMe2SiLi (b) PhMe2SiLi:CuBr,
1:l (c) PhMe2SiLi:CuBr, 2: 1 (d) PhMe2SiLi:CuCN, 2: 1.
resonances at 157.4, ipso; 134.9, ortho; 126.5, meta; 124.6, para; a nitrile carbon at
156.7 and a methyl signal at 5.1 ppm (Figure 22d). As previously reported, evidence for
the formulation of (PhMe2Si)2Cu(CN)Li2 (28) came from the infrared analysis of this
solution which shows a bound nitrile (VCN = 2123 cm-1) but no free L C N . ~ ~ The
coincidence of the 2 9 ~ i chemical shifts of 28 and 27 is attributed to the accidental
chemical degeneracy of the resonances and suggests a similar electronic environment
around silicon in 28 and 27.
Solutions containing PhMe2SiLi (25) and CuBrMe2S in a 3: 1 ratio exhibited a
significant 2 9 ~ i NMR signal at -18.9 ppm as well as a small signal assigned to
(PhMe2Si)2CuLi (27, -24.4 ppm, Figure 21e). An attractive fornulation for the species
exhibiting a 2% signal at -18.9 ppm is (PhMe2Si)3CuLi2 (29). The chemical shift of
this species is identical to that previously observed for (PhMe2Si)3CuLi2 (prepared from
three equivalents of PhMe2SiLi and CuCN, Figure 21e inset). Secondly, the 1 3 ~ NMR
spectra at -70•‹C of the two solutions exhibited identical chemical shifts for silyl bound
methyl and phenyl signals (6 164.3, ipso; 134.5, ortho; 126.0, meta; 122.8, para and a
methyl signal at 8.1 ppm).
That the conversions of "PhMe2SiCuW (38) to (PhMezSi)2CuLi (27) and 27 to
(PhMe2Si)3CuLi2 (29) are reversible was shown by addition of 0.5 equivalent of
CuBrMe2S to 29 whose 2 9 ~ i NMR spectrum is shown in Figure 21e. This resulted in
the regeneration of (PhMe2Si)2CuLi (27, Figure 23a). Further introduction of one
equivalent CuBrMe2S to 27 resulted in the regeneration of "PhMe2SiCu" (38, Figure
23b).
While no free PhMe2SiLi was detectable in solutions conmmng ~5 ana
CuBrMgS in a 3:l ratio, a substantial amount of PhMe2SiLi appeared in the spectrum
of solutions where this ratio was 4: 1 (Figure 210. This interpretation was reinforced by
a positive GiIman for PhMaSiLi (25) in THF and a negative test for all solutions
of 25 containing CuBrMe2S including those where these reagents were present in a 3: 1
ratio. A slight green coloration was observed for PhMe2SiLi:CuBrMe2S in 4: 1 molar
ratio indicating small quantities of fiee PhMe2SiLi (25).
I I 1 1 I I 1 I 1
O -I 0.0 -25.0 -35.0 PPM
Fi y r e 23. 2 9 ~ i NMR spectra in THF at - 5 0 0 ~ (a) (PhMe2Si)3CuLi2 + CuBr, 1:0.5
(b) (PhMe2Si)2CuLi + CuBr, 1: 1.
The relative intensities of the 2 9 ~ i signals attributed to contributing species in
solutions whose spectra are shown in Figure 21 allow estimation of the positions of the
equilibria shown in Scheme 13.
Scheme 13
The equilibrium between PhMe2SiLi (25), "PhMe2SiCuW (38) and
(PhMe2Si)zCuLi (27) lies significantly on the side of 27 (i.e., kl >> k-I). In solutions
comprised of PhMe2SiLi:CuBrMe2S, (3:l) (PhMe2Si)3CuLi2 (29) predominates over
(PhMe2Si)2CuLi by -4: 1. This allows calculation of the equilibrium constant (K2) for
the process. The value of K2 and the error reported were calculated as outlined earlier by
averaging three determinations.36
The ease with which we were able to differentiate between the higher order
silylcuprates derived from CuCN and the lower order reagents prepared from
CuBrMe2S by 1 3 ~ NMR spectroscopy encouraged us to undertake a study of the
complexation of LiX with these reagents. Specifically, solutions containing
(PhMe2Si)2CuLi (27) were prepared from CuBrMe2S or CuI in DMS or THF. The
1 3 ~ NMR spectra were recorded in the region between 0 and 10.0 ppm with the goal of
observing the resonances due to the methyl r d w n c
At -85OC in DMS, the 1 3 ~ NMR spectrum of (PhMe2SihCuLi (27), prepared
from CuBrMe2S and two equivalents of PhMe2SiLi (25, prepared from (PhMe2Si)2
and Li metal)33 consisted of a singlet at 4.2 ppm (in the 0 to 10.0 ppm region, Figure
24a). This was attributed to halide-free (PhMe2Si)2CuLi (27) due" to the precipitation of
LiBr from DMS. Substitution of CuI for CuBr*Me2S gave a solution that exhibited two
signals in the l 3 ~ NMR spectrum: one major (6 4.9) and a minor one at 4.2 pprn (Figure
24b). These were assigned to the methyl carbons of LiI-containing silylcuprate
(PhWSi)2CuLi*LiI (39), and LiI-free silylcuprate (PhmSi)2CuLi (27),
respectively. The assignment of the signal at 4.2 pprn to LiI-free silylcuprate 27 was
facilitated by the observation that the resonance of the methyl in the 1 3 ~ NMR spectmn
of this reagent was precisely at the same position as the methyl resonance of LiBr-free
silylcuprate (PhMe2Si)2CuLi.
The 7 ~ i NMR spectra of (PhMe2Si)2CuLi prepared from CuI and CuBr were in
harmony with the 1 3 ~ NMR results. At -70•‹C in DMS, the 7 ~ i spectrum of the reagent
prepared from PhMe2SiLi and CuBrMe2S (without LiBr) showed a single resonance at
0.43 ppm, which is assigned to the LiBr-free silylcuprate (PhMe2Si)2CuLi (27). Under
the same conditions the 7 ~ i NMR spectrum of (PhMe2Si)~CuLi derived from CuI
consisted of two signals; the major one at 6 0.77 was assigned to (PhMe2Si)2CuLi*LiI
(39) while the minor one (6 0.48) was attributed to the halide-free complex, 27.
The l 3 ~ NMR spectrum of (PhMe2Si)3CuLi2 (29) prepared by the addition of
LiC1-free 25 to LiBr-free (PhMe2Si)zCuLi (27) exhibited signals corresponding to
(PhMe2Si)2CuLi (27,6 4.2) and higher order species 29 (6 5.4, Figure 24c).
Similarly, (PhMe2Si)3CuLi2 (29) prepared from CuI exhibited two major methyl
resonances at 4.9 pprn and 5.4 pprn which were attributed to (PhMe2Si)2CuLi*LiI (39)
and (PhMe2Si)3CuLi2 (29), respectively, along with a minor peak at 6 4.2 pprn which
was assigned earlier to halide-free irn~e2Si)2CuLi (27, Figure 24d).
In agreement with the above results, the 7 ~ i NMR spectrum at -70•‹C of
CuBrMe2S derived (PhMe2Si)3CuLi2 in DMS exhibited signals at 0.45 and 0.2 1 pprn
attributable to 27 and 29 respectively, while (PhMe2Si)3CuLi2 prepared from CuI
1 I I I I I 1
6 5 4 3 2 PPM
Figure 24. l 3 ~ NMR spectra in DMS at -85OC (a) PhMe2SiLi:CuBr, 2:l (b)
PhMe2SiLi:CuI, 2: 1 (c) PhMe2SiLi:CuBr, 3: 1 (d) PhMe2SiLi:CuI, 3: 1.
consisted of a broad multiplet sptmning the region assigned to LiI-containing lower order
species (0.76 ppm) and a halide-free (PhMe2Si)3CuLi2 (29) species (-0.40 ppm). The
presence of broad signals @ these spectra indicates that the rate of exchange between the
various lithium containing species is near the NMR time scale, an observation in
agreement with literature precedents?*9J2
As in the case of preparations of (PhMe2Si)3CuLi2 from CuBrMe2S in THF,
no free PhMe2SiLi was detectable in the solutions of this reagent prepared from CuI in
DMS until the ratio of silyl anion to CuI exceeds 3:l.
In DMS, (PhMe2Si)2CuLi*LiX (39) and (PhMe2Si)zCuLi (27) are in dynamic
equilibria with (PhMe2Si)3CuLi2 (29) as represented in Scheme 14. The reversibility of
interconversion between (PhMe2Si)aCuLi (27) and (PhMe2Si)2CuLi*LiI (39) was
established by the observation that the silyl methyl signals in these species coalesced to a
single peak as the temperature is increased fiom -8S•‹C to O•‹C, and upon cooling to
-85OC, the same-two signal pattern reappears.
Scheme 14
That LiBr is also associated with trialkylsilylcuprates in THF was
evident from 1 3 ~ NMR of (PhMe2Si)2CuLi (27, prepared from two equivalents of
LiC1-free ~hMe2~iLi33 and CuBrMe2S at -85OC). The 1% NMR spectrum
(Figure 25a) showed two methyl signals, the major one at 4.9 ppm and a minor one at
4.2 ppm which are identical to the methyl resonances assigned to (PhMe2Si)2CuLi*LiI
(39) and (PhMe2Si)2CuLi (27) respectively in DMS (Figure 25b).
Ph Me, SiLi : CuBr Me$ 4.9
2:llTHF
I l l I I I
6 5 4 3 ' 2 1 PPM
Figure 25. 1 3 ~ NMR spectra of (a) PhMe2SiLi:CuBr, 2:l in THF (b)
PhMe2SiLi:CuI, 2: 1 in DMS; the spectra were run at -85OC.
It appears that the lower order silylcuprate (PhMe2Si)2CuLi (27) is a more
compler y = r i p c than i t s strai_ghtforward preparation indicates. In DMS, when it is
prepared with Cur, this reagent consists mostly of the LiI-associated copper species,
(PhMe2Si)2CuLi*LiI (39), but when it is derived from CuBr*Me2S, it is free of LiBr.
In THF, when CuI or CuBr.Me2S is used in the preparation, this reagent exists
primarily as LiX-associated copper species and should be represented as
(PhMe2Si)2CuLi*LiX (Scheme 15).
X= I, Br; TFHF 1- - (PhMe2Si)2CuLiLiX (PhMe2SiI3CuLi2
X= Br, DMS (Phh4~Si)~CuLi + LiBr
27
(PhMe$ i)2CuLi. LiI X= I; DMS
39
Scheme 15
One would expect the degree of association of LiX (X = Br or I)
with lower order silylcuprates to vary with the nature of the silyl substitution. We
suspect the trends discovered in this investigation namely that LiX association is high in
solvents such as THF in which LiX salts are soluble to be of general applicability.
It has recently been pointed out by ~ e r t z l 2 that THF and DMS are
complimentary solvents with regard to copper halide based organocuprates. These
reactions commence with copper halides and allcyllithiums and produce lithium halide
which may be associated with the organomprate formed. LiI (the solubility of LiI in
DMS is -10 mdmL at 25OC) and LiBr (the solubility of LiBr in DMS is -0.3 mg/mL at
25OC ) are relatively insoluble in DMS but both LiI and LiBr are soluble in THF
(solubility of LiI and LiBr in THF is >I30 mg/mL at 250~).12 Therefore, in the latter
solvent LiI and LiBr association with cuprates is at least feasible.
The proposed formation of (PhMe2Si)2CuLi*LiX in THF provides an
explanation for the observation that 27 exhibits a 2% NMR resonance with a chemical
shift (Figure 21d) similar to that assigned to (PhMe2Si)2Cu(CN)Li2,28 (Figure 15d).
Thus, if the former existed primarily complexed with LiBr the only structural difference
between 28 and 27 is the substitution of a CN by a Br.
Chemical tests
As observed earlier for silylcuprates derived from CuCN, it was found that
solutions of both 27 and 29 deliver a PhMe2Si group exclusively via 1,4-addition to
cyclohex-2-en-1-one (32, Scheme 16). Solutions composed of 29,25 and CuX in a
Scheme 16. Addition of "PhMe$iCun reagents to cyclohex-2-en-1-one, 32.
1 : 1 :O. 1 molar ratio also added to 32 in a 1,4-manner.32,~5 Similar observations have
been made before in the catalytic cuprate chemistry.
Next the addition of silylcuprates 27,29 and 39 to 1-octyne (35a) was studied.
In agreement with 1 3 ~ NMR xsults, it was found that (PhMe2Si)2CuLi (27, prepared
from CuBr) in DMS added to 35a to yield a mixture of 36 and 37 in a 1: 1 ratio,
whereas addition reactions of (PhMe2Si)2CuLi*LiI (39) and (PhMe2Si)3CuLi2 (29)
Scheme 17. Silylcupration of 1-alkynes.
H H13c& -CH - H13C& -CD +H13C6H
+ H 1 3 c 6 ~ H
(D) H SiMe2Ph PhMe2Si H (D)
% Composition 35b 3 6 3 7 %yield
gave exclusively 36 in -90% isolated yields (Scheme 17). As previously reported,
under similar conditions PhMe2SiLi (25) abstracts the acetylenic hydrogen of 35a to
give 35b as judged by GC-MS (70% incorporation of ZH in 1-octyne using 2 ~ ~ 0 ) ; no
addition products were observed (capillary g.c. analysis).
64
Conclusion
Comparison of the present silylcuprate system with that of the methylcuprate7~8
system reveals several similarities. When the ratio of RLi (R = PhMe2Si or Me) to CuX
(X = I or Br) in THF is 2: 1, a new species, R2CuLi*LiX, is formed in both cases. In the
case of LO Me2CuLi, association of alkyl residues with copper beyond this
stoichiometry does not occur and further addition of MeLi beyond this point gives
solutions containing free MeLi (Scheme I ) . ~
In the case of (PhMe2Si)2CuLi*LiX, addition of further silyllithium gives
solutions which contain negligible amounts of free silyl anions and whose 2 9 ~ i NMR
spectra support the association of three silyl residues with the copper. In THF the novel
species (PhMe2Si)3CuLi2 is farmed regardless of which copper(1) salt is employed.
In DMS, (PhMe2Si)2CuLi (derived from Cur) exists as an equilibrium mixture
of halide-containing (PhMe2Si)2CuLi*LiI (39) and halide-free (PhMe2Si)2CuLi (27)
species. In contrast, lower order silylcuprates derived from CuBr are primarily devoid of
LiBr because of the insolubility of this salt in this solvent. This behaviour is analogous to
that of phenylcuprates.12
MIXED TRIALKYLSILYL AND HOMO- AND MIXED TRIALKYL-
STANNYLCUPRATES DERIVED FROM CUPROUS CYANIDE
Silyl- and stannylcopper reagents are invaluable reagents for the construction of
carbon-silicon and carbon-tin bonds respectively. In general, reactions of these reagents
occur under relatively mild conditions and tolerate polar functional groups. Indeed, one
has many literahm examples from which to decide upon reaction parameters.27,28 Most
commonly sought are lithium-based cuprates, (R3M)nC~Lin-1*LiX (M = Si or Sn, n = 1
or 2, X = Br or CN; 40) prepared stoichiometerically from cuprous salts and 1 to 2
equivalents of ~ ~ ~ ~ i . 2 7 9 2 8 While homoaiallrylsilyl- and stannylcuprates serve
admirably as donors of R3M anions, problems attend the use of these reagents in that not
all the metal anions bound to copper are transferred in these processes. Consequently,
R3MH, R3M-MR3 and R3MOH are produced in workup which complicate product
isolation.27928 . The possibility arises that mixed systems R3Si(R')CuLi*LiX (41) or
R3Sn(R')CuLi*LiX (42) would present opportunities for preferential R3M anion
transfer and thereby increase ligand efficiency of these reagents. The ability of methyl to
serve as an efficient non-transfenable ligand in highly mixed organoalkylcuprates17 led
us to determine the composition of mixed silyl- and stannylcuprates derived fiom CuCN
with methyl serving as the second or third anionic ligand.
It was anticipated that information gained from such studies would not only shed
light into the nature of these cuprates but also provide insights into the selectivity of
ligand transfer. It was envisioned that these studies would help in understanding the
mechanism of catalytic addition reactions of these cuprates and define alternate strategies
for conducting such reactions.
Silicon-29 and Carbon-13 NMR Studies on Mixed TrialkylsilylCuprates
Low-temper& 2% and 1 3 ~ NMR spectroscopy was employed to probe the
composition of solutions generated by mixing dimethylphenylsilyllithium (25) with
equimolar solutions of MeLi and CuCN. Species likely to be formed in these
experiments are higher order mixed silylcuprates PhMe2Si(Me)Cu(CN)Li2 (43),
(PhMe2Si)2(Me)CuLi2 (44) and PhMe2Si(Me)2CuLi2 (45).
Results and Discussion
Dimethylphenylsilyllithium (25) in THF was prepared by reaction of PhMe2SiC1
(30) and lithium metal and therefore contained ~ i ~ 1 . 3 3 Solutions of this reagent at -S•‹C
gave a 2 9 ~ i NMR signal at -28.5 ppm (Figure 26a). Solutions of mixed silylcuprates
(43) were generated by addition of THF solutions of 25 to equimolar THF solutions of
MeLi and CuCN at -70•‹C. The combination of one equivalent of 25 and ~ u ( C N ) ~ i l 7
(16) yielded a solution that afforded a single 2 9 ~ i signal at -20.6 ppm (Figure 26d)
attributable to (PhMe2Si)(Me)Cu(CN)Li2 (43) implying that any equilibrium of the type
shown in Scheme 18 must lie heavily toward 43 (i.e., k p > k-1). To probe the
generality of ligand mobility, solutions containing copper to methyl anion to silyl anion
ratio of 1: 1 :1 were generated by adding MeLi to preformed PhMe2SiCu(CN)Li (26,
Figure 26b). These solutions gave the same spectral results as obtained earlier from the
combination of MeCu(CN)Li and PhMgSiLi.
To establish that 43 could be produced by mutual alkyl and silyl anion exchange.
equivalent amounts of MqCu(CN)Li2'(17) and (PhMe2Si)2Cu(CN)Li2, (28, Figure
26c) were prepared separately and then combined via a cannula at -70•‹C. The 2 9 ~ i NMR
spectrum, taken at -70•‹C, of the reagent formed in this experiment (Figure 26e) was
identical to the one obtained previously (Figure 26d) from the mixture of PhMe2SiLi
(25) and MeCu(CN)Li (16). As anticipated, the spectra from all three reagent
combinations, i.e., MeCu(CN)Li-PhMe2SiLi, PhMezSiCu(CN)Li-MeLi and
Me2Cu(CN)Li2 - (PhMe2Si)zCu(CN)Li2 indicated the same species (43) had been
formed (Scheme 18).
PhMe;?SiCu(CN)Li + MeLi
26 Scheme 18
The 1 3 ~ NMR spectrum (Figure 27d) of (PhMe2Si)(Me)Cu(CN)Li2 (43) at
-70•‹C, prepared from a THF solution of MeCu(CN)Li (16) and PhMe$Wi (Figure
27b),33 consisted of seven lines: four in the phenyl region (6 158.5, @so; 134.9, orrho;
126.7, meta; 124.8, para), one niaile (6 159.4) and two different methyl groups. The
upfield signals were assigned to a methyl bound to silicon (6.0 ppm) and a methyl bound
to copper (-5.0 ppm), respectively, in (PhMe2Si)(Me)Cu(CN)Li2 (43). These results
indicated the presence of a single species, 43, and support the interpretation of the 2%i
NMR spectra (Figure 26d and 26e) given above.
The 1 3 ~ NMR spectra of solutions resulting from the admixture of
Me2Cu(CN)Li2 (17) and (PhMe2Si)2Cu(CN)Li2 (28, Figure 27e) did not show signals
characteristic of the individual reagents (Figure 27a and Figure 27c respectively). Rather
I I I I I I I
0 .la -5.0 -1s . 0 -2s. 0 PPM
Figure 26.2% NMR spectra of (a) PhMe2SiLi (b) PhMe2SiCu(CN)Li (c)
(PhMe2Si)2Cu(CN)Li2 (d) PhMezSiCu(CN)Li + MeLi (e) (PhMe2Si)2Cu(CN)Li2 + Me2Cu(CN)Li2; spectra were run at -50•‹C.
168 154 148 119 8 -1 5 PPM
Figure 2 7 . 1 3 ~ NMR spectra of (a) Me2Cu(CN)Li2 (b) PhMe2SiLi (c)
(PhMezSi)2Cu(CN)Li2 (d) PhMe2SiLi + MeCu(CN)Li (e) (PhMe2Si)2Cu(CN)Li2 + Me2Cu(CN)Li2; spectra were run at -70•‹C.
these solutions gave the same 1% NMR signals as obtained previously from the
combination of PhMe2SiLi (25) and MeCu(CN)Li (16, Figure 27d).
Thus, mixed-alkyl silylcuprates as well as silylcuprates and alkylcuprates,
irrespective of the mode of preparation, revert to the same species, 43;when the ratio of
silyl anion to alkyl anion to cuprous ion is 1 : 1 : 1 (Scheme 18).
The 1% NMR spectra of solutions containing two equivalents of silyl anion to
PhMe$i)2Cu(CN)Li2 + MeLi 2 8
or / and
2 PhMqSiLi 2 5 +
MeLi + .
CuCN
PhMe;!Si(Me)Cu(CN)Li2 + PhMezSiLi
4 3 2 5
or / and
(PhMe$i)2(Me)C~i2 + LiCN 4 4
or / and
+ 0.5 (PhMe2Si)3CuLi2 + LiCN
2 9
or / and
0.66 (PhMe$i)3CuLi2 + 0.33 Me2Cu(CN)Li2 + 0.33 MeLi 4 . 6 6 LiCN
2 9 1 7
Scheme 19
one equivalent each of methyl anion and cuprous ion was next examined. Solutions
generated by addition of two equivalents of PhMe2SiLi (25) to solutions containing one
equivalent each of MeLiFand CuCN would be expected to lead to the formation of an
equivalent each of (PhMe2Si)(Me)Cu(CN)Li2 (43) and PhMe2SiLi (25), or an
equivalent each of (PhMe2Si)2Cu(CN)Li2 (28) and MeLi, or (PhMe2Si)z(Me)CuLi2
(44) and LiCN, or an equivalent each of (PhMe2Si)(Me)2CuLi2 (49,
(PhMe2Si)3CuLi2 (29) and LiCN, depending on whether this mixed system exhibits
behaviour similar to alkyl17 or homosilylcuprates (Scheme 19).
The 1 3 ~ NMR spectrum for the 2: 1 combination of PhMe2SiLi (25) and
MeCu(CN)Li (16) at -85•‹C is shown in Figure 28b. Identical spectra were obtained
when equivalent molar amounts of (PhMe2Si)2Cu(CN)Li2 (28) and MeLi were mixed at
-85•‹C. These spectra exhibited three resonances of unequal intensity in the methyl region
(6 -5.3, -3.7 and 6.0) as well as several signals in the phenyl region (124.7, 126.6,
127.7, 128.6, 134.0, 134.5, 134.9) indicating the presence of more than one species.
Signals at 6.0 pprn and -5.3 pprn are very close to those previously assigned to 43
(Figure 28a), but the absence of signals comsponding to PhMaSiLi (25) rules out the
presence of 43. Futhermore, the absence of signals corresponding to MeLi (-14 ppm)
eliminates (PhMaSi)2Cu(CN)Li2 (28) as a major species and supports our earlier
observation that the silylcuprates exhibit significant tendencies to exist as higher order
cuprates, i.e., (R3Si)3CuLi2.
Warming these solutions to -70•‹C led to more complex spectra (Figure 28a) that
exhibited five peaks in the methyl region (8.1 ppm, 5.9 ppm, 5.5 ppm, -3.8 pprn and
-5.3 ppm).
The composition of the species in solutions comprised of PhMe2SiLi (25), MeLi
and CuCN (2: 1 : 1) was deduced from the absence of (PhMe2Si)(Me)Cu(CN)Li2 (43),
(PhMe2Si)2Cu(CN)Li2 (28), Me2Cu(CN)Li2 (17) and MeLi. The expected species is
(PhMe2Si)2(Me)CuLi2 (44). Thus, the signal at -3.7 ppm was assigned to methyl
bound to copper in (PhMe2Si)2@!g)CuLi2 (44) while the signal at 5.9 ppm was
assigned to the silyl bound methyls in this species. The signals at 8.1 ppm and 164.3
ppm correspond to the methyl and ipso carbon, respectively, in (PhMe2Si)3CuLi2 (29).
Observation of (PhMe2Si)3CuLi2 (29) requires species such as
8 . k n 9 ,,
(Si) MeCu
I- I 1 L I
168 154 138 9 - 8 -1 5 PPM
Figure 2 8 . 1 3 ~ NMR spectg of (a) 2 PhMe2SiLi + MeLi + CuCN at -70•‹C (b) 2
PhMe2SiLi + MeLi + CuCN at -8S•‹C (c) PhMezSi(Me)Cu(CN)Li2 at -70•‹C.
(PhMe2Si)(MehCuLi2 (45) with two akyl residues per copper atom to maintain methyl
anion/copper cation balance. Thus, the peak at -5.3 ppm was assigned to the two copper
bound methyls of PhMe2Si(MebCuLi2 (45) and the signal at 5.5 ppm is attributed to
the silyl bound methyls in this species. In agreement with this assignment the signal at
-3.7 ppm assigned to 44 decreases in intensity with respect to that at -5.3 ppm assigned
to 45, as the solution is warmed, and produces signals due to 29.
According to this analysis, the mixed silyl-alkylcyanocuprates wherein the silyl
anion to alkyl anion to copper cation ratio is 2:l:l exist in the dynamic equilibria shown
in Scheme 20. The equilibria are similar to those observed earlier in the case of
homosilylc yanocuprates.
+ 3 MeLi 1-
+ 1 3 CuCN
Scheme 20
F (PhMe2Si)$u(Me)Li2 4 4 +
PhMt$3(Me)2CuLi2 45 +
(Pl~hie~Si)~CuLi~
2 9 +
L 3 LiCN
Addition of three equivalents of PhMe2SiLi (25) to a combination of one
equivalent each of MeLi and CuCN could result in the formation of one or all of the
possible combinations of silylcuprates shown below in Scheme 21.
3 PhMe,SiLi 2 5 +
MeLi +
CuCN
(PhMe2Si)2Cu(Me)Li2 + PhMe2SiLi + LiCN 44 2 5
or / and
- (PhMe2Si)3CuLi2 + MeLi + LiCN
2 9
or / and
0.5 PhMe,Si(Me)2CuLi2 + PhMe2SiLi + 0.5 (PhMe2Si)3CuLi2 + LiCN 45 2 5 29
Scheme 21
The 1 3 ~ NMR spectrum of the solution generated from the addition of 3.0
equivalents of 25 to one equivalent of MeLi and one equivalent of CuCN at -85OC
showed sirnil& features to the ones generated fiom the combination of 2.0 equivalents of
dimethylphenylsilyllithium to one equivalent of methyllithium and one equivalent of
copper cyanide. The signal originally present for MeLi (-14 ppm) was replaced by
signals characteristic of the mixed cuprates obtained previously for the 2:l: 1 case, i.e.,
(PhMe2Si)2(Me)CuLi2 (44, (PhMe2Si)(Me)2CuLi2 (45) and (PhMezSi)3CuLi2 (29).
Similar results were obtained by addition of MeLi to preformed (PhMe2Si)3CuLi2 (29)
at -85•‹C. This experiment shows that for this ratio of silyl anion to cuprous ion,
PhMe2SiLi (25) is liberated by methyllithium.
Thus, solutions generated by combination of PhMezSiLi, MeLi and CuCN in the
ratios of 2:l:l and 3:l:l contain 44,45 and 29 except that in the latter the proportion of
PhMgSiLi is increased in accordance with the equilibria shown in Scheme 22. In
support of this analysis, the 2 9 ~ i NMR spectrum of solutions generated from admixture
of 3.0 equivalents of PhMe2SiLi (25) and one equivalent each of MeLi and CuCN
showed appreciable amouhts of free PhMe2SiLi (25, -28.5 ppm).
9 PhMe$iLi 25 (PhM%Si),Cu(Me)Li2 + PhM~si (Me)~CuLi~ + + 4 4 45
+ 3 PhMqSiLi + ( P h M ~ s i ) ~ C u L i ~ + 3 LiCN 3 CuCN 25 29
Scheme 22
Tin-119, Carbon-13 and Hydrogen-1 NMR Studies on
Homotrialkylstannylcuprates
Prior to examination of the composition of mixed triakylstannylcuprates
R3Sn(R')CuLi=LiX (42), the composition of homotrialkylstannylcuprates was studied.
Low-temperature 1~ NMR spectroscopic studies were initially conducted on
(Me3Sn)nCuLin-1*LiCN (n = 1,2 or 3) in the region between 1.0 ppm and -2.0 ppm
with the specific goal of observing the resonances due to the various methyl groups.
Reaction of (Me3Sn)z (46) and methyllithium at -45OC in THF gave a solution
containing equimolar amounts of MeqSn (6 0) and MegSnLi (47).46a,b These
preparations gave a broad 1~ signal at -0.35 ppm (Figure 30a)46c.d and, as previously
reported, no l~(119sn-lH) coupling was observed. The 119% NMR47 spectrum of
solutions resulting from the reaction of hexamethylditin (46, Figure 3 la) with MeLi
revealed a signal for Me3SnLi (47, Figure 31b) at -187.9 ppm. The high field position
of this signal (the highest'of any trimethyltin derivative yet reported) has been attributed
to the presence of the negative charge on the tin [i.e., (~e3~n)- ] .48 The l 3 ~ { 1 ~ } NMR
spectrum of solutions of -Me3SnLi (47, Figue 32a) in THF exhibited a broad signal at
-3.7 ppm. Contrary to what might be expected of this solution, no 13~-119sn coupling
was observed for this species in THF. At least two mechanisms can account for this lack
of Sn-C coupling. First, rapid exchange of methyl groups between Me3SnLi and
(Me3Sn)z would destroy the correlation between the l 3 ~ and l 9 ~ n spin states and
therefore no coupling would be seen. Second, association of THF molecules to give
[Me3Sn(THF)X]- solvated anions would lower the symmetry of the anion and
quadrupolar relaxation effects would average out the coupling. At present we cannot
distinguish these two possibilities, although the lack of coupling even at -8S•‹C, where
methyl exchange rates are apt to be slow, suggests that solvation of the anion is a more
likely explanation. This explanation is consistent with the observation that the 1~ and
1 3 ~ NMR chemical shifts of this reagent are concentration dependent.
Solutions of stannylcuPrates28 we= generated by addition of copper0 cyanide
to THF solutions of Me3SnLi (47) at -78OC. Combination of equimolar ratios of 47 and
CuCN resulted in a nearly homogeneous solution, which showed several 1~ NMR
signals in the range -0.1 to -0.3 ppm with major signals centered near -0.18 ppm (Figure
30b). The major signal at -0.18 ppm is assigned to polymeric Me3SnCu(CN)Li (48).
Supporting evidence for this formulation comes from infrared analysis of these solutions
which show a major absorption due to bound nitrile at 21 11 cm-1 accompanied by a less
intense absorption at 2082 cm-1 due to free L ~ c N . ~ ~ The presence of the latter requires
Me3SnLi (47) be present. However, the 1 3 ~ NMR (vide infra) failed to reveal this
component. We attribute the presence of free LiCN as detected by infrared spectroscopy
as being due to thermal decomposition of Me3SnCu(CN)Li (48) during the infrared
analysis.3k
-~ -
-6 0.0 -5 PPM
Figure 29. 1~ NMR spectra of (a) Me3SnLi (b) 1.0 Me3SnLi + CuCN (c) 1.5
Me3SnLi + CuCN (d) 2.0 Me3SnLi + CuCN (e) 2.5 Me3SnLi + CuCN (f) 3.0
MegSnLi + CuCN (g) 4.0 Me3SnLi + CuCN; the spectra were run at -70•‹C.
The 1: 1 combination of Me3SnLi and CuCN yielded a slurry exhibiting a
multitude of l9sn signals in the region of -140 to -160 ppm accompanied by a minor
signal at -109 ppm due to MegSn2 (46, Figure 31c). The former were assigned to
polymeric Me3SnCu(CN)Li (48) based on the observations made earlier on
PhMe2SiCu(CN)Li (vide supra). The 1~ NMR spectrum of this solution exhibited
several broad signals centered around -0.18 ppm supporting the view that this reagent is
polymeric.
The solution of "Me3SnCuW (49) prepared from Me3SnLi (47) and CuBrMe2S
exhibited several 1 1 9 ~ n signals centered around -179 ppm (Figure 3 Id) suggesting it is
also polymeric. In agreement with this interpretation most of the intensity of 1 3 ~ and 1~
NMR signals was lost in solutions containing equirnolar amounts of 47 and
CuBrMe2S. The signals that were visible were due to (Me3Sn)a (46). Lower order
cuprates containing one equivalent of RLi to each cuprous ion equivalent such as
PhMe2SiCu(CN)Li, "PhMe2SiCuW and C H ~ C U ~ , ~ are also polymeric. Appearance of
signals for MegSn2 (46) in the 1 1 9 ~ n NMR spectrum but not in the 1~ or 1 3 ~ spectra
(vide infra) is attributed to the decomposition of Me3SnCu(CN)Li (48) during the longer
time required to acquire the l 1 9 ~ n spectrum.
The 1 3 ~ NMR spectrum of 48 exhibited a broad triplet at 4.5 ppm (Figure
32b). The possibility that the multiplicity of this signal is due to 2~ ( 6 ~ i - 1 3 ~ ) coupling
( 6 ~ i and 1 3 ~ coupling has been observed in M ~ ~ c u L ~ F ~ 1) is ruled out because the
observed satellites (total intensity of 16.5%) are more intense than calculated (7.42% of
6 ~ i ) and the coupling is significantly larger than expectede41 The satellites of this signal
are attributed to lJ(119sn-13~) coupling (Figure 32b) in agreement with natural
abundance of l 9 ~ n (natural abundance of 119sn is 8.5% and that of 17sn is 7.6% ;
since the ratio of ~ 1 1 7 s n - C / ~ 1 1 9 ~ n - ~ = 1.046, individual satellite peaks due to each of
these isotopes cannot usually be resolved for MegSnM species47a). The very low
u 0 -100 -200
PPM
u 0 -100 -200
PPM
Figure 30. 11%n NMR spectra of (a) Me3SnSnMe3 (b) Me3SnLi (c) 1.0 Me3SnLi + CuCN (d) 1.0 Me3SnLi + CuBr, the spectra were run at -70•‹C.
-1 5.0 PPM
Figure 31. 1 3 ~ NMR spectra of (a) Me3SnLi (b) 1.0 Me3SnLi + CuCN (c) 2.0
Me3SnLi + CuCN (d) 3.0 Me3SnLi + CuCN (e) 3.0 Me3SnLi + CuBr; the spectra were
run at -70•‹C.
coupling constant of J = 63 Hz is indicative of a trimethyltin group bonded to a weakly
e l e c t r o n - a . g moiety as in the case of Me3SnLi [ l ~ ( l '9sn-l3c) = 120
~~)],46c,d,47a,49a -
As the ratio of Me3SnLi (47) to CuCN was increased from 1: 1 to 2:1, a new
peak at -0.24 ppm in the 1~ spectrum appeared and increased in intensity at the expense
of the signal at -0.18 ppm (Figure 30c). When the ratio of Me3SnLi (47) to CuCN
reached 2:1, minor signals due to Me3SnCu(CN)Li (48,6 -0.18) and (Me3Sn)3CuLi2
(50,6 -0.27, vide infra) and a major signal at -0.24 ppm (Figure 30d) appeared. The
latter signal is attributed to a species containing Me3Sn/Cu ratios between 1: 1 and 3: 1.
It is significant that both the 1~ and 1 3 ~ (vide i@a) NMR spectra revealed the
absence of 47 in these solutions. This limits consideration of the composition of the
stannylcuprates possessing Me3SdCu ratios between 1 : 1 and 3 : 1 to (Me3Sn)3CuLi2
(SO), (Me3Sn)2Cu(CN)Li2 @I), (Me3Sn)~CuLi (52) and (Me3Sn)3Cu2Li (53,
Scheme 23).
0.33 (Me3Sn)3Cu2Li + 0.33 (Me3Sn)3CuLi2 + LiCN
f 5 3 50
/, M e , S n C T + Me3SnLi 4 7
2 Me3SnLi + CuCN or
47 (Me3Sn)2Cu(CN)Lb
5 1 or
(Me3Sn)2CuLi + LiCN
Scheme 23
The 1 3 ~ NMR spectra of solutions prepared from 2.0 equivalents of Me3SnLi
(47) and one equivalent of CuCN contained three signals (6 1.67, -0.04 and -1.13,
Figure 32). Each signal- is accompanied by a set of satellites attributed to coupling of the
two tin nuclides of spin 1/2 to carbon. The observation of spin coupling allows the
establishment of Cu-Sn stoichiometry in solution47a based on the relative intensities.
Thus, the signal centered at 1.67 pprn exhibited 1 1 9 ~ n - 1 3 ~ coupling of 44.3 Hz and
satellites of 45% intensity indicating 1 3 ~ coupling to three tin atoms, while that at -0.04
pprn was accompanied by satellites revealing a l % n - l 3 ~ coupling of 46.8 Hz to two
tin atoms (26%) and the signal at -1.13 pprn exhibited 1 1 9 ~ n - 1 3 ~ couplings of 46.8 and
23.8 Hz due to two different tin atoms in the ratio of 3:2.
The signal at 1.67 pprn is assigned to (Me3Sn)3CuLi2 (50) based on the
observation that this signal grows at the expense of the signals at -0.04 pprn and -1.13
pprn as the Sn/Cu ratio is increased from 2:l to 3:l and beyond (Figure 32c, 32d and
3%). The presence of signals carresponding to 50 requires the presence of species
containing a 1 5 1 ratio of Me3Sn/Cu to maintain the copper-tin balance in these
solutions. The 1% NMR spectrum of a solution generated by admixture of 1.5
equivalents of Me3SnLi (47) with one equivalent of CuBrMe2S exhibits a broad peak
at -1.13 pprn Hence, (Me3Sn)3Cu2Li (53) is assigned to the latter signal. The l 3 ~
NMR spectrum of a solution generated from the combination of two equivalents of
Me3SnLi (47) and one equivalent of CuBroMe2S exhibits a broad signal at -1.35 pprn
assigned to (Me3Sn)2CuLi (52). This signal is absent in solutions generated from a 2: 1
combination of Me3SnLi and CuCN. Thus, the stannylcuprates eliminated from
consideration as the species giving rise to the signals centered at -0.04 pprn are
Me3SnCu(CN)Li (48), (Me3SnhCuLi (52) and (Me3Sn)3Cu2Li (53). This signal is
therefore attributed to (Me3Sn)2Cu(CN)Li2 (51).
This interpretation is in apement with the 1~ NMR analysis of these solutions
which revealed a major signal at -0.24 pprn at a Sn/Cu ratio of 2:l and that the intensity
of 1 1%n-13~ satellites surrounding the 1 3 ~ NMR signal for this reagent (vide supra)
corresponded to the presence of two tins.
Sequential addition of Me3SnLi (47) to the above solution (< 0.5 equivalent)
resulted in the appearance of a new signal at -0.27 pprn in the 1~ NMR spectrum (Figure
30e). When the ratio of 47 to C u m was precisely 3:1, the major 1~ signal observable
was at -0.27 pprn accompanied by a minor signal due to (Me3Sn)zCu(CN)Li2 (51,
Figure 30f). The peak at -0.27 pprn was attributed to (Me3Sn)3CuLi2 (50) by analogy
with our earlier observations on the ttiall<ylsilylcuprates.
In support of our interpretation of the 1~ NMR spectral data, the 1 3 ~ NMR
spectrum of this solution showed a major signal at 1.67 pprn along with a minor signal
assigned to 51 and 53 (Figure 32d). Moreover, solutions of Me3SnLi:CuBrMe2S
(3:l) exhibited l 3 ~ NMR spectra with a signal at 1.67 pprn (Figure 32e).
While no appreciable amounts of Me3SnLi (47) were detectable by 1~ NMR
spectroscopy in solutions containing 3.0 equivalents of Me3SnLi to one equivalent of
CuCN, substantial amounts of Me3SnLi were detected by ZH NMR analysis (Figure
30g) in solutions containing more than 3.0 equivalents of 47 to each equivalent of
CuCN.
That the formation of Me3SnCu(CN)Li (48) and (Me3Sn)3CuLi2 (50) are
reversible was shown by addition of 0.5 equivalent of CuCN to the solution whose 1~
NMR spectnun is shown in Figure 30f. This results in the regeneration of the spectrum
shoxz k Z;ig&"~ 3Cd. Pc!knr %troduction of CuCN (1.0 equivalent) to this solution
results in the regeneration of the spectrum shown in Figure 30b. Thus, the species
exhibiting signals at -4.5 ppm, -1.13 ppm, -0.04 pprn and 1.67 pprn are in dynamic
equilibria as represented by Scheme 24.
Me3SnLi + CuCN ,- Me3SnCu(CN)Li
Scheme 24
The chemical shift change obsewed when lower order
stannylcyanocuprates are converted to higher order stannylcyanocuprates is opposite to
what would be expected from simple arguments based on electronegativity. Coordination
of an electron-rich Me3SnLi (47) to the copper centre in Me#nCu(CN)Li (48) should
increase the electron density at copper and hence at both tins in the resultant HO species.
The 1195, resonance in the HO species would therefore be expected to be upfield of that
in the LO reagents, not downfield as found. An anomolous pattern of shielding was also
observed for 2% in aiallrylsilylcyanocuprares (46) and presumably has its origin in the
p-orbital "imbalance" and its contribution to the paramagnetic term of nuclear
rh&aldincr 37b ---..- ---- 0
The observed 1J(l l g ~ n - 1 3 ~ coupling decreases as one proceeds from LO to HO
stannylcuprates i.e., 1 ~ ( l l g ~ n - l 3 ~ , 63 Hz) of Me3SnCu(CN)Li (48,S -4.5) >
1 ~ ( 1 1 9 ~ n - 1 3 ~ , 46.8 Hz) of (Me3Sn)2Cu(CN)Li2 (51, S -0.04) > 1~(119~n-13~, 44.3
Hz) of (Me3Sn)3CuLi2 (50,S 1.67). The small 1 1 9 ~ n - 1 3 ~ coupling observed in HO
stannylcuprates as compared to LO stannylcuprates is consistent with a greater electron
density on the tin atom in the former.46dv48 According to this rationale a greater negative
charge on the tin atom would lead to a greater s-character in the orbital containing the
lone pair bound to copper.46d An increase in the fractional s-character of the bond
should lead to a decrease in s-character of the remaining tincarbon bonds and result in a
lower 1 1 9 ~ n - 1 3 ~ coupling as observed.46d947a
A final point of interest is the observation that the intensities of the 1 1 9 ~ n - 1 3 ~
satellites surrounding the 1 3 ~ NMR signals at 1.67 ppm, -0.04 ppm and -1.13 ppm
(attributed to (Me3Sn)3CuLi2 (50); (Me3Sn)2Cu(CN)Li2 (51); and (Me3Sn)3Cu~Li,
(53); respectively) with respect to the central signal are greater than the 15:85 ratio
expected for 1 3 ~ coupling to a single 1 1 9 ~ n nucleus. We have observed the same
phenomenon for stannylmetalloids such as Me3SnALEt2 (54) and
[(Me3Sn-9-BBN*OMe)-]Li+ (55). The small coupling constdnts of 54 [ 1 ~ ( 1 1 9 ~ n - 1 3 ~ ,
40 Hz] and 55 [ 1 ~ ( 1 1 9 ~ n - 1 3 ~ , 56 Hz] are very close to those observed for the
stannylcuprates mentioned above.
A priori one would ascribe the increased satellite intensities to a rapid exchange
of akyl residues between tins. At least in the case of stannylcuprates we have
demonstrated that such exchange does not occur between Bu3Sn and Me3Sn groups
associated with cuprous ion. Thus, one equivalent of CuCN was added to one equivalent
each of Bu3SnLi and Me3SnLi at -70•‹C and the reaction quenched with NH&1 after two
hours at -50•‹C. Analysis by GC-MS revealed no Bu2MeSn or BuMe2Sn moieties.
Although this experiment rules out the exchange between methyl and a butyl group on tin
in HO stannylcuprates, it does not unequivocally rule out methyl-methyl exchange in
these cuprates. Crossover experiments using cuprates generated from (CH3)3SnLi and
(CD3)3SnLi should be canied out to observe the formation of (CD3)n(CH3)3-nSn
species in the reaction ~ X I ~ X N ~ S .
An alternative possibility is that the enhanced satellite intensity is due to the
equivalency of lJ(119sn-13~) and 3 ~ ( 1 1 9 ~ n - 1 3 ~ ) in these complexes. For alkyl tin
compounds where the 1%n-13~ coupling is through a carbon frame, 1~ > 3~ > 2~ > 4~
and all these can be appreciable.50a Thus, the equivalency of 1~ and 3~ in the present
system is not entirely unreasonable.
Tin-119, Carbon-13 and Hydrogen-1 NMR studies on Mixed
Trialkylstannylcuprates
We next focused on mixed reagent (Me3Sn)(Me)Cu(CN)Li2 (56). Earlier
spectroscopic studies on the silylcuprates (vide supra) indicated that
(PhMe$3)(Me)Cu(CN)Li (43) can be produced either by the sequential addition of
MeLi and PhMe2SiLi (25) to CuCN or upon admixture of Me2Cu(CN)Li2 (17) with
(PhMe2Si)2Cu(CN)Li2 (28). If (Me3Sn)(Me)Cu(CM)Li2 (56) is produced in similar
experiments with stannyl anions one should obtain two high field signals comsponding
to two methyl groups, one bound to copper and the other to tin. On the other hand the
various equilibria exhibited by the higher order homo-trimethylstannylcuprate (51, vide
supra) could result in the formation of species having more than one tin.
The 13c NMR spectrum of a solution containing one equivalent each of MeLi
and Me3SnCu(CN)Li, 48 (Figure 33b) at -70•‹C is shown in Figure 33d. The
combination of these reagents resulted in a simple spectrum exhibiting three signals at
-2.06 ppm, -9.32 pprn and -9.5 pprn (MeqSn). The peaks at -2.06 pprn and -9.32 pprn
were assigned to carbons of the methyls bound to tin and copper respectively, in
(Me3Sn)(Me)Cu(CN)Li2 (56). The same spectrum (Figure 33e) was obtained when
equimolar amounts of 17 (Figure 33a) were combined with 51 (Figure 33c). The
intensities of the satellites of the signal at 2.06 pprn are as expected for 1 ~ ( 1 ~ 9 ~ n - ~ 3 ~ ,
182.0 Hz) coupling to a single tin nucleus. No 2 ~ ( 1 1 9 ~ n - 1 3 ~ ) coupling was obsetved
between the methyl and the Me3Sn moiety. This is attributed to a rapid intermolecular
exchange between methyls on copper. Since neither (Me3Sn)2Cu(CN)Li2 (51),
MezCu(CN)Li2 (17), Me3SnCu(CN)Li (48), Me3SnLi (47), MeCu(CN)Li (16) nor
MeLi are seen when the Me3SnLi:CuCN:MeLi ratio is exactly 1 : 1 : 1, an equilibrium
favouring 56 (Scheme 25), analogous to Scheme 18, must lie far to the products.
(Me3Sn)&u(CN)Li2 + Me2Cu(CN)Li2 1 F 51 1 7 (Me3Sn)(Me)Cu(CN)Li2
5 6
1, Me3Sn(Me)zCuLi + LiCN
5 7
Me3SnCu(CN)Li + MeLi 4 8
Scheme 25
-1 5 PPM
Figure 32. 1% NMR spectra of (a) Me2Cu(CN)Li2 (b) 1.0 Me3SnLi + CuCN (c) 2.0
Me3SnLi + CuCN (d) Me3SnCu(CN)Li + MeLi (e) (Me3Sn)2Cu(CN)Li2 + MaCu(CN)Li2; the spectra were run at -70•‹C.
Me, SnU '1.1
PPM
I Me,Sn(Mo) Cu CN U ,
0.0 -1.0
PPM
PPM
L I W 0 -GO -120 -200
PPM
ngure 33. IH NMR spectra of (a) Me3SnLi (b) 1.0 (Me3Sn)2Cu(CN)Li2 + Me2Cu(CN)Liz; NMR spectra of (c) Me3SnLi (d) Me3SnLi + MeCu(CN)Li; the
spectra were run at -70•‹C.
(Me3Sn),Cu(CN)Li,: MeLi
Figure 34. 1 3 ~ NMR spectra of (a) (Me3Sn)(Me)Cu(CN)Li2 (b)
(Me3Sn)(Me)Cu(CN)Li2 + MeLi (c) (Me3Sn)3CuLi2 + (Me3Sn)(Me)Cu(CN)Li2 (d)
2.0 (Me3Sn)2Cu(CN)Li2 + MeLi; the spectra were run at -70•‹C.
In support of our interpretation of the 1 3 ~ NMR spectrum of 56, the IH
spectrum of this reagent (Figure 34b) at -70•‹C (prepared from equimolar amounts of
(Me3Sn)2Cu(CN)Li2 (51) and Me2Cu(CN)Li2 (17) showed two resonances in a 3:l
ratio at -0.45 (Me on tin) and -1.56 pprn (Me on copper) along with a minor peak due to
51 at -0.24 ppm. Similar results were obtained from the combination of Me3SnLi
(Figure 34a), MeLi and CuCN. The 1 1 9 ~ n NMR spectrum (Figure 3 4 ) of 56 showed a
single peak at -182 pprn also indicating a single species.
Addition of one equivalent of MeLi to preformed (Me3Sn)(Me)Cu(CN)Liz (56,
Figure 35a, containing MeqSn, -9.5 pprn), at -70•‹C, yielded a solution that exhibited two
1 3 ~ NMR signals at -1.7 pprn 1 ~ ( 1 1 9 ~ n - 1 3 ~ , 189 Hz) and -9.3 pprn which were
attributed to (Me3Sn)(Me)2CuLi2 (57, Figure 35b). This assignment was based on the
increase in the intensity of the signal at -9.3 pprn when the ratio of Sn/Cu/Me was 1 :1:2.
No signals corresponding to other stannylcuprates were apparent in these solutions
(Figure 35b). Also, no 2 ~ ( 1 1 9 ~ n - 1 3 ~ ) was observed in this cuprate due to rapid
intermolecular exchange between methyls on copper.
Addition of one equivalent of (Me3Sn)3CuLi2 (50) to 56 (2:1:0.5, Sn:Cu:Me)
resulted in the appearance of signals for free Me3SnLi (47) as well as signals attributable
to (Me3Sn)3CuLi2 (50), (Me3Sn)3Cu2Li (53) and (Me3Sn)(Me)2CuLi2,57 (Figure
3%). Similar results were obtained when 1.0 equivalent of MeLi was added to 2.0
4 (Me3Sn)2Cu(CN)Li2 + 2 MeLi
Me3Sn(Me)2CuLi2 + Me3SnLi + (Me3Sn)3CuLi2 + (Me~jSn)~Cu~Li + 4 LiCN
5 7 4 7 50 5 3
Scheme 26
equivalents of preformed (Me3Sn)2Cu(CN)Li2 (51, Figure 35d). Appearance of
Me3SnLi (47) in the latter experiment indicates the most basic ligand, in this case methyl
anion, is preferentially bound to copper in solutions &ficient in methyl anion (Scheme
26).
Encouraged by the formation of (Me3Sn)(Me)Cu(CN)Li2 (56) we examined the
low temperature 1% NMR spectra of solutions generated from the mixture of one
equivalent of Bu3SnI-I (58) and Me2Cu(CN)Li2 (17). This combination resulted in the
appearance of a single peak at -13.1 pprn (Figure 36a) corresponding to the methyl in
Bu3Snm (59, Scheme 27). The stoichiometry of this reaction requires the formation of
Me(H)Cu(CN)Li;? (60) which has recently been detected by Lipshutz in 1,4-addition
reaction of the hydri&.50a
Scheme 27
Addition of a further equivalent of Bu3SnH (58) to the above solution resulted in
immediate release of a gas and appearance of a new signal at -9.3 ppm (Figure 36b).
This signal is very close to the one observed earlier (-9.2 ppm) for methyl bound to
copper in (Me3Sn)(Me)Cu(CN)Li2 (56). Hence, we attribute this peak to methyl bound
in @u3Sn)@@Cu(CN)Li2 (61). The stoichiometry of the conversion of an equivalent
of Bu3SnH (58) and Me(H)Cu(CN)Li2 (60) to produce (BugSn)(Me)Cu(CN)Li: (61)
requires the evolved gas to be hydrogen (Scheme 27).50b
Figure 35. 13c NMR spectra of (a) Bu3SnH + Me2Cu(CN)Li2 (b) 2.0 Bu3SnH + MaCu(CN)Li2; the spectra were run at -70•‹C.
On the basis of these NMR investigations, it is clear that ligand exchange in
lower order and higher order cuprates occurs between silyl, stannyl and alkyl anions.
Because of these equilibria silyl- or stannylcuprates and alkylcuprates readily form mixed
metallo-alkylcuprates. For solutions containing equimolar ratios of R3M (M = Si or Sn)
anion, methyl anion and cuprous ion, the mixed cuprate RsM(Me)Cu(CN)Li2 is the
predominant species in solution (Scheme 28).
As the amount of R3M anion is increased, a species containing two R3M anions
to each methyl and cuprous ion [(R3M)~(Me)CuLi21 is formed. [R3M(Me)2CuLi2] and
(R3M)3CuLi2 are also formed in these solutions. This shows that equilibria among these
three species is close to unity. The observation that R3MLi is present in solutions of
higher order metallocuprates provides a pathway by which ligand exchange on cuprous
ion can occur [i.e., via free R3MLi(R'Li)].
The ability of methyl ligands to bind tenaciously to cuprous ion can be explained
by the differences in basicity of MeLi (pKa do), R3MLi (pKa =23, M = Si or Sn) and
"R3MCu" (pKa =15).2801 This implies that in the mixed "R3Si(Me)Cun or
"R3Sn(Me)CuW reagents [i.e., (PhMe2Si)(Me)Cu(CN)Li2 (43);
(Me3Sn)(Me)Cu(CN)Li2 (56); (Me3Sn)(Me)$uLi2 (57) and (Bu3Sn)(Me)Cu(CN)Li2
(61)] there would be a tendency for the silyl- or the stannyl- ligands to transfer faster
than methyl anions. This would result in the conservation of the more valuable silyl or
stannyl residue which would, in turn, increase the efficiency of R3M transfer reactions.
Product yields would also be expected to improve because of simplification in isolation
procedures since methane is the hydrolysis product whereas R3MH and R3MOH are the
side products obtained on workup of homo-trialkylsilyl- or stannylcuprates.
R3MLi 0.66 (R3M)3CuLi2-- - -- - - +
0.33 MefluCNLi2 + 0.33 MeLi + 0.66 LiCN
I
- - ,
v 0.5 (R3M)3CuLi2 4
I
I (R3M)2(Me)CuLi2 R3MLi + I I + I
0.5 R3M(Me)2CuLi2 I I LiCN I
I
+ I I
I I
LiCN i R ~ M L ~ I I I I
I I
I I
I I
1 I I
I I I
I I I
I I I
1 I I
I : R3MLi I
I I I
i I I i I
R3MLi 0.33 (R3M)3CuLi2- - - 4 - - - - - - - - - - - - - - - (R3M)2Cu(Me)Li2 + I
I + 0.33 (R3M)(Me)2CuLi2 + I
I R3MLi + LiCN I
v / 0.33 (R3M)2(Me)CuLi2 + I I 0.
R3MLi + LiCN I 0
I 0
0 . I 0 . I 0
0
. I . I 0
I 0
0*
t 0.5 (R3M)3CuLi2- - - - - - - - - - - - - - - - - - - - - - - .(R3M)3CuLi2 + MeLi
+ + 0.5 R3M(Mez)CuLi2 LiCN
+ R3MLi + LiCN
Scheme 28
Reactions of Silyl- and Stannylcuprates with a,S-unsaturated Ketones
Evidence has beeil obtained from our 1 3 ~ NMR studies (vide supra) that in the
mixed cuprates of the general formulae (R3IvQn(Me)CuLin-l*LiCN (n = 1 or 2), both
Me and R3M moieties move rapidly between parent species to preferentially yield the
mixed ligand metallocuprates. It was therefore of interest to evaluate ligand transferability
from these reagents to organic substrates such as a , ~ n o n e s as a step toward more
efficient synthetic methodology. Futhermore the simple spectroscopic behaviour of the
mixed ligand metallocuprates afforded an opportunity to study the mechanism of the
reaction of these reagents with a$-unsaturated ketones.
The facility with which cuprates bearing alkyl,2 silyl?2 and stan11~1~~ anions
introduce these groups in a Michael k s e to a&unsaturated carbonyls compounds
(62) makes them the reagents of choice for these transformations. Most mechanistic
studies of these processes have involved Gilman reagents (i.e., RzCuLi, 2). While initial
coordination of the substrate carbonyl with lithium (2 + 62 -, 63) and formation of a
substituted enolate (64) are common features, (Scheme 29) there has been considerable
divergence of opinion concerning the intervening steps.
Kinetic data52 on these Michael reactions as well as isolation of an insoluble
species53 that was convertible to a substituted ketone (implying that the unknown
species was the enolate, 64153 suggest a process involving intermediates that
unimolecularly rearrange to the enolate (64, Scheme 29). The correlation of reduction
potentials of a&unsaturated carbonyl systems with cuprate reactivity has given rise to a
proposal involving a single electron transfer mechanism.% Initial Lewis acid-Lewis base
interaction (63, Scheme 29, step al) encourages transfer of an electron from a dimeric
cuprate to the enone to give 65 (Scheme 29, step a2) followed by formation of a copper-
carbon bond in the species 66 (Scheme 29, step ag). Reductive elimination (Scheme 29,
step a4) fkom the species 66 affords the enolate 64, although the exact nature of
M in the species 64 is still an open question. Intermediate 66 can also arise by way of an
initial charge transfer complex (67, Scheme 29, step b l -+ b l or via b3 -+ a3).55 Still
more direct would be the nucleophilic addition of the reagent to the P carbon atom of the
substrate56 without the intervention of a single electron transfer step (Scheme 29, step
a1 -+ cl -+ cz). Direct carbocupration to give an a-cuprioketone (69, Scheme 29, step
d l or via steps a1 -+ cl -+ dz) followed by rearrangement to the thermodynamically
more stable lithium enolate (Scheme 29, step d3) is another altemative.57
The notion of an early intermediate complex (beyond that of simple Li+
coordination) has gained considerable momentum since its original formulation. Infrared
spectroscopic studies of reactions involving unsaturated esters58 and, in particular, low-
temperature 1~ and 1 3 ~ NMR studies of the reaction of Gilman's reagent, ~ e 2 C u ~ i 5 9
(1) and of the mixed lower order cuprate ~ e ( 2 - t h . i m ~ l ) C u ~ i ~ with cinnamatk esters
(70) provide cogent evidence for binding between copper and the n* MO of the enone.
Thus, addition of 70 and 1 in THF-dg at -70•‹C gave an intermediate species in which the
1 3 ~ labelled C2 carbon shifted upfield by 55 ppm to give a quartet of signals at 53.7,
52.6,51.0 and 49.9 ppm which were ascribed to an intermediate n-~om~lex.60,6~
These shifts are in the range of coordination shifts (- 60 ppm) for trigonal alkene-
transition metal complexes of Ni, Pd and Pt.62~63 Since diastereoisomerism would be
expected to yield only two complexes (i.e., only two signals) the additional signals
must be related either to E, Z isomerism, solvent association or aggregation state.59960
The study described here involved measurements of the 1 3 ~ NMR spectra of
solutions generated during the addition of metallocuprates to cyclohex-2-ene-1-one (32)
at low temperature with the aim of observing resonances due to the copper containing
species formed following the delivery of R3M (M = Si or Sn) from
(R3M)n(Me)CuLin-l*LiCN to 32. The logical byproduct of ligand transfer is the LO
cyanocuprate MeCu(CN)Li (16) which should be observable over time.
When reactions of 32 with PhMe2Si(Me)Cu(CN)Li2 (43) or
MegSn(Me)Cu(CN)Li2 (56) were monitored by 1 3 ~ NMR spectroscopy at -85OC the
appearance and gradual increase of a peak at -13.5 pprn corresponding to 16 was readily
discerned.There was also evidence for the involvement of a complex between higher
order silyl- (29 and 43) as well as HO stannylcuprate (56) and cyclohex-2-ene-1-one
(32).
The 1 3 ~ NMR spectra of the cuprates and cyclohex-2-ene- 1 -one are complex.
Some features, however, are prominent and facilitate interpretation. These are
particularly the methyl signals and the signals of the C2 carbon of the enone as illustrated
in Figure 36a. On addition of (PhMe2Si)3CuLi2 (29) at -85OC to 32 the resonances for
C2 (120.9 ppm) and C3 (150.7 ppm) of 32 disappear and several closely spaced signals
are generated that are initially centered around 60 pprn (59.2 and 61.7 ppm) and around
35 pprn (32.7 and 33.8 pprn). Coincident with these new signals, four major (-2.37, -
2.69, -4.4, -5.14 ppm) and two minor (3.9,4.0 ppm) resonances attributable to silyl
bound methyls appear. As the solution is allowed to stand at -70•‹C, the high field methyl
resonances at -4.4 and -5.14 pprn disappear while those at 3.9 and 4.0 pprn grow. This
is accompanied by a decrease in the signal intensities near 60 and 35 pprn and an increase
at 38.6 and 23.7 ppm. The signals originally obtained for the nitrile carbons (159.4 ppm)
are replaced by another at 152 ppm. Workup of this reaction yields 33 (Scheme 30). It is
attractive to attribute the signals centered at 35 and 60 pprn to C2 and C3 of a copper-
alkene-n-complex (68, Scheme 29, R = PhMe2Si) while the signals at 38.6 and 23.7
pprn are attributed to the C2 and C3 of the enolate 64 (Scheme 29, R = PhMeaSi). The
signal due to the C3 carbon of the product appears at 26 ppm. The upfield methyl shifts
at -2.37 and -2.67 and -4.4 and -5.14 are assigned to the silicon bound methyls in the
a-cuprioketone, 69 (Scheme 29) and the copper-alkene-n-complex, 68 respectively
(Scheme 29, R = PhMe2Si). The signal due to the carbonyl (C1) is difficult to
distinguish from the baseline noise due to long relaxation decay of the quaternary carbon.
Similarly, the low-temperature 1 3 ~ NMR spectrum of PhMe2Si(Me)Cu(CN)Li2
and 32 shows four major signals at -2.37, -2.67, -4.40 and -5.14 pprn and two minor
signals at 3.9 and 4.0 pprn (Figure 36b). The important feature is the appearance of
signal at -13.5 pprn corresponding to MeCu(CN)Li (16) as a result of exclusive silyl
group transfer from the silylcuprate to cyclohexenone. The appearance of 16 suggests
that under these conditions, some addition of 43 to 32 has occurred. It also allows one
to deduce that lithium is coordinated to the enolate product, 64 (Scheme 29, M = Li).
One set of doublets (3.9 and 4.0 ppm) may be attributed to the silyl bound methyls in the
common enolate adduct (64, Scheme 29). The corresponding signals for C2 of the
adduct appear around 40 pprn while those for C3 absorb at 24 ppm. These signals are
close to that expected for the silyated enolate corresponding to 64 (R = PhMe2Si). The
h ~ d signal for the nitrile (159.4 ppm) is replaced by another at 15 1.3 pprn as the
reaction progresses. The other remaining set of signals around 60 and 35 pprn are
hypothesized to be due to a copper-alkene x-complex (corresponding to the adduct 68,
Scheme 29).59-61
The formation of a chiral center on mixing a silylcuprate (29) with 32 is a
straight-forward basis for the presence of two signals for silyl bound methyls in the 1 3 ~
NMR spectrum corresponding to the enolate 64 (3.9 and 4.0 ppm) and the copper-
alkene n-complex G8 (-4.4 and -5.14 ppm; Scheme 29). The presence of lithium salts
PPM Figure 36. 1 3 ~ NMR spectra of (a) (PhMe2Si)3CuLi2 + 3.0 equiv 32 (b)
PhMgSi(Me)Cu(CN)Li2 + 32; the spectra were run at -8S•‹C.
in the solutions of these cuprates may be responsible for the appearance of additional
signals at -2.4 and -2.7 ppm. Coordination with these salts may result in the formation of
a stable adduct 69 (Scheme 29) since no signals were obtained around -2.0 ppm in
solutions free of LiC1. Alternatively, the ,aggregation state of the silylcuprates and the n-
complexes could vary and hence give rise to species the enone carbons in different
magnetic environments.
Similar results were obtained by using the HO homocuprate 28 and the HO
stannylcuprate 56. In the latter case signals near 35 ppm along with a doublet centered at
12 ppm were produced on initial mixing at -85OC and these signals disappeared as signals
for stannylated enolate (38 ppm) and MeCu(CN)Li (6 -13.5) appeared.
On the basis of this preliminary spectroscopic information, it is proposed that
conjugate addition of mixed metallocuprates to a$-unsaturated ketones involves initial
formation of an intermediate copper(1)-olefin-n-complex 68 which is stable at very low
temperatures. In the presence of lithium salts appreciable amounts of a-cuprioketone
(69) is also produced. Warming the solutions leads to exclusive delivery of the metallo
anion with simultaneous formation of the LO alkylcyanocuprate.
In accordance with our NMR results, it was found that solutions of
PhMe2Si(Me)Cu(CN)Li2 (43), Me3Sn(Me)Cu(CN)Li2 (56) and
Bu3Sn(Me)Cu(CN)Li2 (61) deliver a PhMe2Si or R3Sn (R = Me, Bu) group
Scheme 30. Reaction of metalocuprates with 32.
exclusively via 1,4-addition to cyclohex-2-en-1-one (32) to give 3-silyated (33) or 3-
stannylated (70) cyclohexanones respectively (Scheme 30).32y46 3-Methyl-
cyclohexanone which would be produced from 1,6addition of a methyl group was not
observed.
Thus, the transferability of R3Si or R3Sn anions bound to copper is higher than
that of the methyl li&nds. The preference for R3M (M = Si or Sn) transfer presumably
reflects the kinetic reactivities of higher order mixed metallocuprates, as well as the
stability of the resulting lower order reagent formed [i.e., Me(Cu)CNLi].
Conclusion
These studies indicate that higher order cuprates containing silyl or stannyl and
alkyl anionic ligands (e.g., R3Si(R1)CuLi-LiX (41) and R3Sn(R')CuLi*Li (42) are . similar to akyl cuprates (17) with respect to their tendencies to readily exchange ligands
between copper centers. These studies also present evidence that such mixed cuprates
transfer silyl and stannyl ligands in preference to alkyl ligands in reactions with a$-
enones. The formation of a copper(1)-alkene complex as an initial step of the conjugate
addition of homo-trialkylsilylcuprates, homo-trialkylstannylcuprates as well as mixed
stannyl- or mixed silylcuprates to enones has been demonstrated but additional work
using a 1 3 ~ labelled enone is required to firmly establish the suggested parallel.
Final comments
One of the overriding questions regarding the nature of metallocuprates is the
aggregation state of the solution species. If one uses a c h i d ligand R* to prepare
R*(R3M)Cu(CN)Li2 then if the R* is optically pure and the species is monomeric one
should obtain signals for the 1 3 ~ and 1~ spectra of the species (i.e., same 6 for R and
S). If the R* is optically pure and the species is dimeric, a single signal should appear for
each magnetically nonequivalent atom. If the R* is racernic and the species is monomeric
then one signal should appear for each magnetically nonequivalent atom. If, on the other
hand the species is dimeric then atoms magnetically equivalent in the monomer would
appear nonequivalent in the dimer and give more than one signal due to the
diastereoisomeric species that are possible (different signals for RR, SS and RS and
SR).
IN SITU GENERATION OF ACTIVATED METALLOCUPRATES
During the evolution of stoichiometric lithium-based stannyl- and silylcuprates in
synthetic chemistryF7-32 one aspect of their chemistry that has been invariant is their
mode of preparation.27J8 Both lower order and higher order stannyl- and silylcuprates
are routinely prepared fiom &(I) salts and 1 or 2 equivalents of R3MLi (M = Sn or Si).
When the Cu(1) salts are cuprous halides 2.0 equivalents of R3MLi are required to
induce ate formation. When CuCN is used direct ate formation occurs. This contrasts
with organometallic chemistry outside the organocopper arena where preformed
organometallic complexes are commonly produced via ligand exchange between metal
centers. 1
As reported in the previous section ligand exchange [e.g., R2&(CN)Li2 and
(R3M)2b(CN)Li2, M = Si or Sn] readily occurs between silylcopper or stannylcopper
derivatives and alkylcopper systems. The equilibria established by the many
combinations examined led to the view that exchange between anionic organic ligands
such as R3Si, R3Sn and alkyl associated with cuprous ion is indeed facile. On the basis
of these observations, it was decided to examine the chemistry associated with ligand
migration on copper where ligands with different electropositive metals are involved.
Specifically, it was envisioned to prepare derivatives of trialkylstannyl and
trialkylsilyl anions in which the counter ion is a metal-containing species capable of
exerting an effect as an electrophilic additive. An example of the type of reaction we
hoped to achieve is shown in Scheme 3 1.
R3Si-M(R), + CuX - [R3SiCu(X)] M(R),
7 4 7 5
Scheme 31
These reactions were of interest because, if successful, they would allow the in
situ generation of activated stannyl(73) and silylcuprates (75) directly from
trialkylstannyl- (72) and trialkylsilyl metalloids (74) respectively, thereby bypassing the
generation of thermally labile stannyl- or silylcuprates. Mixed ligand cuprates such as 73
and 75 would, by analogy to activated alkylcuprates, possess increased reactivity toward
organic substrates. If the stannyl and silylcuprates are indeed activated by the presence of
the electrophilic M(R)n group, the literature did not give an indication of this at the onset
of our work.
Certainly, one of the recent and most encouraging aspects of organocuprate
chemistry has been the improvement made in reactivity when other species are added to
the reactions of lower or higher order alkyl~u~rates.6 Phosphorus-containing
compounds (e.g., phosphines and phosphites, etc.) have been utilized for decades to
solubilize the cuprous(1) salts and also to act as stabilizing ligands. Recently, however,
the use of electrophilic species such as BFyEt20, Et2AlC1, ZnBn, MgBr2, T i c 4 and
Me3SiC1 with Gilman cuprates (R2CuLi, 2) has significantly expanded the scope of this
methodology.6 In general, these electrondeficient reagents tend :G ;iii:i6iuc. LUG, &G
and yields of cuprate reactions. Since no changes in protocols for cuprate use are
necessary, aside from the introduction of additive, it is not surprising that the fxequency
of their use continues to grow?$
Although commonly used, little is actually known about the role of these Lewis
acids in the reactivity modification. It is presumed that these additives function by Lewis
acid-base complexation with the substrate while the cuprate remains unperturbed.
However, it has been recently shown by low-temperature NMR spectroscopy that the
addition of Lewis acids such as BFyEt20 and Me3SiC1 leads to formation of modified
c~~rates.64365 Thus, addition of BFyEt2O to a solution of Me2Cu(CN)Li2 (17) leads
to immediate association of the nitrile ligand of the cuprate with the additive. It also
sequesters MeLi from the cuprate cluster to form MeLioBF3 and the lower order cuprate
MeCu(CN)Li (16, Scheme 32).64
Scheme 32
Similarly, Me3SiC1 reacts with the higher order cuprate 17 to give Me3SiCN and the
corresponding lower order cuprate 1 (Scheme 33).65
Me&u(CN)Li2 + Me3SiC1 - Me3SiCN + LiCl + Me2CuLi
1 7 1
Scheme 33
Our reasons for expecting the desired ligand exchange (Scheme 3 1) to occur lay
in the strong tendency of organodirnetallics containing coordinatively unsaturated metals
to form bridged complexesl (76) through three-center-two-electron bonds. Where
participating metals are identical, such associations are of little synthetic consequence.
When the metals are different, however, it can lead to transmetallation, ate formation or
complexes in which the charge has been nversed on 2~ compared to that in ute
formation (Scheme 34).1
When the electronegativities of 1~ and 2~ are substantially different, the more
electronegative metal is reported to amact both bridging ligands, i.e., 1~ and 2L,
inducing ate formation as shown in 77. Ate formation increases the nucleophilicity of
the Z M ~ L moiety, as in the conversion of organocoppers2 to lower order, ( R ~ c u L ~ ) , ~ ~
or higher order, ( R ~ c u ( c N ) L ~ ~ ) , ~ P ~ , ~ cuprates. Ate formation is also responsible for
increasing the nucleophilicity of carbon-metal adducts during the conversion of
organoboron and organoaluminum compounds into organoborates66 and
organoaluminates.67 It can also suppress undesirable electrophilicity of the 2~ center,
such as the B and A1 atoms of organoboranes and organoalanes, respectively.l
Alternatively, transmetallation can result from dissociation of the bridged species
76 into two new coordinatively unsaturated species.l?68 This is a very general process
in organometallic chemistry.
Finally, polarization69 within 76 through threecenter-two-electron bridging of
the two metal centers can occur as shown in 78. The 2~ center in 78 is both
coordinatively unsaturated and positively charged. Such metal centers are more
electrophilic than in the unpolarized monomer.
Since the types of species (73 and 75 ) expected to be produced in Scheme 31
were logical intermediates in the cuprous ion catalyzed reaction of stannyl and silyl
metalloids with 1-alkynes, we adopted the reaction in Scheme 35 as a test of the
aforementioned strategy of in situ activation.
80a
Scheme 35
CU(I) MEDIATED METALLOMETALLATION OF 1-ALKYNES
The fikt synthetically useful metallometallations of 1-alkynes were those
involving the addition of the stannylcopper reagent 49 to 1-alkynes (Scheme
36).28c,30e-h
MeOH "9 - Me3Sn Cu.SMe2 Me3Sn "9 H
8 1 8 3
MeOH ------- x* Rg H SnMe3
82
Scheme 36
These reactions give products possessing the regiochemistry shown in 83 almost entirely
devoid of 84.308 The presumptive initial adducts (81 and 82) were reported to require
consumption of the vinyl-copper center by in situ protonolysis, presumably to overcome
an unfavorable adduct + alkyne equilibrium?8~30g Attempted capture of the
intermediate vinyl-copper compounds with electrophiles other than proton was
unsuccessful.30~-h
Parallel with the development of stannylcupration, addition of silylcuprates (28)
to 1-alkynes was investigated (Scheme 37).30a-d In the resulting adducts (85 and 86)
subsequent capture of the vinyl-copper bond with both proton and carbon electrophiles
was achieved and yielded vinyl silanes (36 >> 37) possessing regiochemistry opposite
to that observed in the case of ~tann~lcu~rations.30a-d
Scheme 37
Regiochemical control is a challenge because additions of Si-Cu and Sn-Cu
reagents to 1-alkynes are reversible.28830g Thus, the first step in Schemes 36 and 37
is reported to be thermod'ynamically unfavomble, and the stannylcupration (Scheme 36)
is reported to proceed to completion only when an alcohol is present in situ to hydrolyze
the vinyl carbon-copper bond in the intermediate 81.3% Assuming a reversible reaction,
the regiochemistry in metallocupration reactions would probably be dictated by the
greater ease of protodecupration of 81 and 85 than of its regioisomer 82 and 86
respectively. Thus, other factors must be at work to explain why stannylcupration yields
2-stannylated alkenes (83) form 1-akynes whereas silylcupration yields 1-silylalkenes
(36).
Metallometallation of 1-akynes has been developed as an extention of stannyl-
and silylcuprations of 1-alkynes (Scheme 38, shown for stannylmetallation only). Such
reactions have been sucessfully extended to bimetallic reagents derived from Sn-
~1$8b,c30 s ~ - M ~ ? O S n - N O S n - ~ n ? l Sn-~,48872 Sn-~i,73 Sn-Sn,74 Si-~1?5
S i - ~ ~ , 7 5 Si-~n?5~76 si-Si77 and Si-~n.71,78 Additions are very slow without added
catalysts; Cum and Pd(0) complexes have emerged as the most effective catalysts in
terms of yields and regiochemical bias. In the case of Cu(1) catalysis one can postulate
79a and 87a as the initial adducts. For Cu+ to behave as a catalyst, subsequent reaction
of the vinyl carbon-copper bonds in 79a and 87a must occur to give the products (79b
and 87b, respectively) of transmetallation by M(R)n. The reactivities of these final
adducts (79b and 87b) require electrophilic consumption of the more reactive
organometallic center prior to isolation and yield 88 and 89 respectively. Reaction of the
second vinyl organometallic center with an electrophile commonly involves electrophilic
addition in case of silicon79 or transmetallation in case of tin.80 Thus, the utility of
metallometallation lies in the simultaneous generation of two stereo- and regiodefined
vinyl organometallics of differential reactivity.
Several problems beset the stannyl and silyIrnetallations of 1-alkynes shown in
Table m. Reactions utilizing reagents wherein M(Rh = ~1,70975 ~ ~ 7 0 9 7 5 and
Zn74975976 are reported to require a three fold excess of reagent to achieve high alkyne
consumption. The use of excess reagent leads to the formation of by-products which are
Scheme 38
114
Table 111. Addition of Silyl- or Stannylmetalloids to 1-alkynes.
Reaeent Catalvst om %vield 88a 89a
THF, O•‹C
m, O0C
THF, O0C
TMEDA,OQC
TMEDA,O•‹C
m, O•‹C MeOH
60-70•‹C 48h
85OC
THF, 25OC
THF, 0•‹C
m , O0C
THF, O•‹C MeOH
m , O•‹C
THF, 25OC
THF, 25OC
THF, O0C
not easily separated from the vinylsilane30&~v46 or stannane48 products.
Regiospecificity is high for some stannylmetallations (Scheme 38, Table III) 88a/89a,
S n - g O (95/5), Sn-Si73 (90110) but low for others Sn-~170 (62/38), s ~ - M ~ ~ O (30/70),
and S n - ~ 7 2 (35165). Only for the Sn-Zn and Sn-Si cases is the regiochemical bias
synthetically useful. Control of regiochemistry in the addition of R3Si-M(Rln and
R~SII-M(R)~ reagents to 1-alkynes is a goal that has been pursued only for silicon-zinc
reagents. Nozalci and co-workers75 reported that the Cu+ catalyzed addition of
PhgSiZnEt~Li to 1-alkynes gave only regioisomer 89a. A related reagent76 possessing
larger zinc akyl groups, PhMe;!SiZntBu2Li, added to 1-alkynes to give almost
exclusively (>99: 1) the alternate regioisomer, 888.76
Proposed work
It is proposed to study the scope and mechanism of copper catalyzed addition of
stannyl and silylmetalloids to 1-alkynes with an aim to:
i) determine if metallocuprates containing electrophilic cationic metals are
activated towards elemphilic consumption of the vinyl copper bond in the expected
adducts.
ii) achieving regiochemical control.
iii) determining the synthetic potential and utility of the expected vinyl metal
bonds in the adducts.
iv) determine the compatibility of metallometallations with polar functional
groups.
Applications of the developed chemistry to the synthesis of pheromomes.
Effect of Method of Preparation of Trialkylstannylithium on the
Efficiency of Stannylalumination and Stannylboronation
Capture of R3SnLi by methyl iodide provided a facile assay of the efficiency of
R3SnLi preparation (Table IV). Reaction of B U ~ S ~ H ~ ~ with LDA (entry 5)49a at low
temper- and reaction of Me3SnSnMe3 with MeLi (entry 6) were clearly the most
efficient.3w
Table IV. Efficiency of Production of R3SnLi Observed Upon Quenching with Me1
Entry Reactants and Conditions 5% yield of By-Products(% yield).
THF, O•‹C, 15min. BugSnCl(15%).
2 Bu3SnCl, Li (wire) 48 BugSn2 (30%).
THF, O•‹C, 24h. ~ u q ~ n (30%), Bu3SnCl(4%).
3 MqSnCl, Li (dis) 70 M e n 2 (30%).
THF, O•‹C, 8h.
4 Bug SnH, BuLi, 4 BuqSn.
THF, O•‹C, 15rnin.
5 Bug SnH, LDA, 90 BugSn;! (8%).
THF, -30•‹C, 15min.
6 Me3SnSnMeg. MeLi on u-d
-40•‹C, THF, 20min.
Reaction of Bu3SnLi with Et2AlCl in THF gave solutions of Bu3SnAlEt2 (90)
which, in the presence of Cu+ salts, reacted with ldecyne (91) to give, after
protonolysis, the vinyl staxmanes 92a and 93a (Scheme 39, Table V). In situ addition
of a proton source was not necessary to achieve high conversion of alkyne to product in
these reactions. Indeed, the more efficient the reaction yielding Bu3SnLi, the more
efficient the subsequent stannylalumination of 1-decyne (Table V, compare entries 1 and
2; 3 and 4; and 5 and 6). Vinyl stannanes (92a and 93a) were separated by preparative
gas chromatography and their structures deduced by IH, 1% and 1 1 9 ~ n NMR
Electro~hile HCl 92a, E = H 90% 93a, E = H 2 ~ ~ 1 92b, E = 2~ 92% 93b, E = 2~
CH&HCH213r 92c, E = CH2CH=CH2 89% Me1 92d, E = Me 76% 93d, E =Me PhCH2Br 92e, E = CHgh 84% PhI 92f, E = Ph 79%
spectroscopy as well as GC-MS. Evidence for the structure and stereochemistry of 92a
and 93a was provided by the magnitude of the coupling constants (~Js.-H - 140 Hz for
92a and - 70 Hz for 93a) between the 117~11 and 1 1 9 ~ n isotopes and lH. These values
are typical of vinyl stannanes having trialkylstannyl groups rmzs andcis , respectively to
vinyl hydmgens.50 Cis addition to the alkyne was confixmed by the disappearance of the
high field70 vinyl hydrogen signal in the 1H NMR spectrum of 92b when the reaction
of 90 (Bu3SnAlEt2) with 91 (I-decyne) was quenched with 2 ~ ~ l . Regioisomer 93a
was prepared independently by hydroalumination67 of 91 with DIBALH,
transmetallation (n-BuLi) and reaction of the alkenylalanate with Bu3SnC1 (Scheme 39).
Proton magnetic resonance and mass spectra as well as the gas chromatographic retention
time of this sample were indistinguishable from those of 93a obtained by
metallometallation (Scheme 39).
Effect of Catalyst on Stannylalumination
Use of PdO, pd2+ or Cu+ complexes as catalysts resulted in efficient addition of
BugSnAlEt2 (90) to ldecyne, 91 (Table V). Addition of 91 to THF solutions of 90 at
-30•‹C in the presence of Cu+ salts resulted in high yields of vinyl stannanes 92a and
93a with a synthetically useful regiochemical bias favoring 92a. Use of CuCN gave
higher yields and higher regiochemical bias than CuBr-Me2S (Table V, compare entry 2
with 8). Regioisomer 93a was favored (90: 10) when CUI was used as the catalyst
(compare entry 2 with 4). Catalysts based on PdO generally gave mixtures of 92a and
93a which were rich in 93a (Table V, entries 5-7). Yields and regiochemical bias were
lower for palladium catalysts than for CuCN.
Table V. Addition of Bu3SnAlEt2 to 1-I3ecyne.a
BuqSnBu6Sr
B u ~ S ~ M ( R ) ~ Catalyst %yield 92a+93a Entry Bu$MM(Rh Allryne (solvent) Tanp 9 2 a 93a (alkyne unreacted)
-30•‹C
-30•‹C
-30•‹C
-30•‹c
OOC
OOC
-30•‹c
-wc
OOC
0%
-30•‹C
-30•‹C
-30•‹C
-30•‹C
-30•‹C
-30•‹C
aSee Scheme 39 for numbering. b ~ u 3 ~ n ~ i prepared from SnC12 and BuLi. CBu3SnLi prepared from
Bu3Sn.H and LDA. d ~ a l f of the theoretical amount of alkyne was added, followed by CuCN (5 mol %).
then the remaining amount of alkyne.
Effect of the Mode of Addition of Reagents on Stannylalumination and
Stannylboronation
A solution of 91 was added to a cold (-30•‹C) THF solution of 90 followed by
the Cu+ catalyst; it was then quenched with 1M HCL This solution yielded vinyl
stannane products (92a >> 93a) of higher regiochemical purity than when the reaction
was conducted by adding solutions of 90 to 91 (Table V, compare entry 11 with 12,13
with 14, and 15 with 16).When the m e was added in one portion to the
organometallic reagent it was necessary to use excess tin reagent to achieve high alkyne
consumption flable V, entries 2,11 and 13). The excess reagent was eventually
converted to hexaalkylditin and tetraallcyltin. Slow addition of alkyne to 1.2 equivalents
of 90 at -30•‹C resulted in high alkyne consumption and minimal formation of
hexabutylditin (Table V, entry 16). Excess 1 - m e can presumably provide a proton,
which can react with the intermediate generated. Normant has recently shown that slow
addition of 1-alkynes to organometallics ar low temperature improved yields of
carbocupration reactions.82
Similarly, copper (I) catalyzed addition of [(Bu3Sn-9-BBNaOMe)Li+] (94) to
1-nonyne (95), followed by quenching with 1M HCI yielded vinyl stannanes (96a and
97a) with high regiochemical bias (Scheme 40) in favour of 96a. Use of ~ H C I as the
quenching agent gave high yields of 96b, confirming the syn mode of addition of 94 to
the alkyne (95). Use of CuBraMe2S gave higher yields and higher regiochemical bias
than CuCN. Regioisomer 97a was favoured (60:40) when phenyl acetylene was used as
the a l h e . As observed for stannylalumination (vide supra) yields and regiochemical
bias were lower for palladium catalysts than for CuBraMe2S.
i) CuBr - 9 4 + ii) E h m ~ h i l e E SnBu3
9 6 9 7
Electro~hile 9 5
HC1 96a, E = H 92%
PhCH2Br 96e, E = CH2Ph 78% PhI 96f, E = Ph 63%
Scheme 40
. Effect of Solvent on Stannylalumination and Stannylboronation
Addition of polar aprotic solvents such as dimethylformami& (DMF) or
dimethylsulphoxide (DMSO) had no effect on the course of Cu(1) catalyzed
stannylalumination. Addition of HMPA, surprisingly, reversed the regiochemistry of the
reaction. The latter reactions were conducted by adding HMPA to cold THF solutions of
Bu3SnLi followed by addition of Et2AlC1, ldecyne (91) and CuCN. After
consumption of 91 ceased the reaction was quenched with 1M HCl to yield vinyl
stannane 93a (Scheme 39) as the major regioisomer (94:6) (Table V, compare entry 1
with 9 and 2 with 10).
Compatibility of Polar Functional Groups with S tannylalumination and
Stannylboronation
That stannylalumination and stannylboronation are compatible with polar
functional groups was shown by the efficient reaction of Bu3SnAlEt2 (90) and
[(Bu3Sn-9-BBN*OMe)-Li+] (94) with 5-hexyne-1-ol(98a), 5-hexyne-1-acetate (98b),
1 -(tetrahydro-pyrany1oxy)-5-hexyne (98c), 1 -bromo-5-hexyne (98d) and 1 -c yano-5-
hexyne (9%) in the presence of Cu(1) catalysis to yield vinyl stannanes 99a - e (Scheme
41). In case of 1-bromo-5-hexyne, a product arising from intramolecular cyclization
resulting from the formation of the other regioisomer was also obtained. The only other
tin containing product in these reactions were hexabutylditin which was easily separated
90, M(R), = A1Et2 X(CH2)4- + Bu$nM(R),
9 8 a , X = O H b, X = OAc 1 i) CuCN c,X=OTHP ii) HCl or d, X = Br ally1 bromide e , X = C N
Scheme 41
by chromatography on silica. Furthermore, in situ protonolysis of the presumptive 12-
dimetallo adduct, 79a or 87a (Scheme 38) is not necessary to achieve high
consumption of alkyne if a reactive coupling reagent is added to the reaction.83 Thus,
addition of 94 to the functionalized alkynes (98a-e) with CuBroMe2S as the catalyst,
followed by the addition of one equivalent of cuB~Me2S as the coupling agent and allyl
bromide as the electrophile,8 gave decent yields of 100a-e with only minor
contamination of the other regioisomer thereby making the entire process extremely
versatile (Scheme 41).
Reactions of 1,2-cis-Dimetallo-1-Alkenes with Electrophiles.
The dirnetallic adducts generated in the reaction of Bu3SnAlEt2 (90) with 91
underwent either transmetallation with n-BuLi or Pd(0) catalyzed cross-coupling
reactions exclusively at the vinyl-aluminum bond (Scheme 39). For instance,
stannylalumination of 1-decyne catalyzed by CuCN, followed by transmetallation of the
vinylalane moiety with n-BuLi resulted in the intermediate alanate. Reaction of the alanate
with excess allyl bromide84 in THF or methyl iodide in HMPA gave good yields of 92c
and & N ~ S of 926 and 934 respectively. Addition of 3 mole% of Pd(PPh3)2C12-2.0
DIBALH (diisobutyl aluminum hydride)85 to the adduct derived from the addition of 90
to 91 under CuCN catalysis followed by addition of allyl bromide, benzyl bromide or
iodobenzene86 gave excellent yields of 92c, 92e or 92f in high stereo- and
regiochemical purity (Scheme 39).
Likewise, addition of 1-nonyne (95) to cold THF solutions of [(Bu3Sn-9-
BBN0OMe)-Li+] (94) followed by CuBrmMe2S and use of allyl bromide87a and ethyl
vinyl ketone in the presence of BFyEt20 (1.0 equivalent) as the electrophile gave
modest yields of 96c and 96d respectively (Scheme 40). Addition of the mixture
resulting from reaction of 94 and 95 to a solution of 5 mole% of Pd(PPh3)4 and ally1
brornide,87b benzyl bromide,87c iodobenzene87d as the electrophiles and NaOMe as the
base gave 96c, e-f respectively in high regio and stereochemical purity (Scheme 40).
No products arising from cross-coupling reactions of vinylstannane were observed.
Synthesis of Trisubstituted Alkenes.
Vinylstannanes 92c, 92d and 93d were further reacted with electrophiles under
transmetallation conditions to afford olefins lOla - i (Schemes 42 and 43). In each case
the reactions resulted in cross-coupled products derived from retention of configuration
with respect to the vinyl-tin bond. In the case of 92c (Scheme 42), reaction with I2 in
CH2C12 smoothly produced the vinyliodide, lola, which underwent facile metal-
halogen exchange with excess n - ~ u ~ i . g g Addition of Me3SiC1 gave lOlb in good yield
i) n-BuLi 1 TMEDA -60•‹C + RT
H17C* Bu3Sn ii)Me3SiCI
92c 1Olb
I / CH2C12 i) 2.2 eq. n-BuLi -%IT -7g•‹C, 0.5 h
ii) Me3SiC1
I / i) 2.2 eq. n-BuLi -7g•‹C, 0.5 h
ii) D20 D
lOla l O l c
Scheme 42
whereas quenching the reaction with 2 ~ 2 0 gave an excellent yield of 101c. Olefin
lOlb was also synthesized in moderate yield by treatment of 92c with n-BuLi /
T M E D A ~ ~ ~ followed by trapping of the vinyl anion with Me3SiC1. The low yield of
product in this reaction is attributed to poor transmetallation due to steric congestion of
the txibutylstannyl group.73,89
Oxidative addition of either benzyl bromide or ally1 bromide at Pd(Ph3P)4
followed by coupling with vinylstannane 92d (Scheme 43) in refluxing THF gave
excellent yields of 101d and 101e, respectively.gO The vinyl stannane 93d also
underwent facile coupling with (a-1-iodo-hexene (101f) under Pd(0) catalysis to yield
stereochemically pure 1,3diene, 101g.91 Palladium catalyzed crosscoupling of
Pd(Ph3P)2 prepared from reaction of Pd(PhsP)2Clz with two equiv. of DIBALH
Scheme 43
the iodide derived from 93d with 1-hexynl-n-tributylstannane (101h) cleanly gave
stereo-defined 13enyne; l0li.92 The rate of the coupling reaction was sensitive to
catalyst. High ratios of phosphine ligand to palladium slowed the reaction.92 Hence, the
coordinatively unsaturated Pd(0) catalyst derived from reduction of (Ph3P)2PdC12 with
2.0 equivalents of D I B A L H ~ ~ was used for crosscoupling reactions during the
synthesis of lOlg and 101i.
Synthesis of the pheromone of the Square-Necked grain beetle
The square-necked grain beetle, Cathartus quadricollis , is a cosmopolitan pest of
stored grain products. In North America this shiny, reddish-brown beetle is abundant
chiefly in the Southern United States and infests a variety of stored commodities such as
corn and peanuts. The beetle is also found in the wild and attacks a large variety of plant
seed pods. In morphology and habit, C. quadricollis is similar to other grain-infesting
beetles of the genera Cryptolestes and Oryzaephilis but is larger and strong flyer.93 The
major differences between C. quadricollis and the other cucujids studied to date is that
the aggregation pheromone of this species appears to be a single component, and is
assigned the shucmre 7-methyl-(622)-nonen-3-yl acetate (102, Scheme 44).94 It was
envisioned that addition of bimetallic reagents to 1-butyne followed by sequential
quenching of the vinylbimetallic adduct with Me1 and ethyl vinyl ketone should result in
the formation of the keto- intermediate in a single step. Reduction followed by acetylation
would give the desired product of correct stereo and regiochemistry.
Thus, addition of PhMe2Si(Me)Cu(CN)Li2 (43) to 1-butyne (103) followed by
electrophilic capture of the vinylsilylcopper intermediate with Me1 in HMPAllYF
resulted in the formation of 104.46a ~odonation-transmetallation79 followed by
conversion of the vinyllithium to the vinylcopper species and addition to ethyl vinyl
ketone yielded the ketone which was reduced to the alcohol and then to the pheromone
(102, Scheme 44) in two-steps in overall yield of 53%. The same precursor was also
synthesized from the addition of 94 to 1-butyne followed by addition of ethyl vinyl
ketone in the presence of BFyEt20. Transmetallation of the vinylstannane and
subsequent coupling with Me1 completed this sequence.
12 n - B a , CuBr
Scheme 44
Mechanistic Studies
In the previous section, it was demonstrated that under Cu(1) catalysis bimetallic
reagents derived from Sn or Si add efficiently to 1-akynes. These additions yield
organometallics that contain two carbon-metal bonds of differential reactivity. The
chemistry ensuing from this methodology has yielded powerful new methods for
stereospecific synthesis of trisubstituted olefins under mild conditions and in the
presence of polar functional groups.48 Although regiochemical control of metallo-
metallations was possible, the mechanistic features of this reaction are not fully
understood.
In this section we describe chemical and spectroscopic studies aimed at
elucidation of the mechanism of the addition of trialkylstannyl- and silylmetalloids to
l - a w e s . Specifically, we examined the 2H, 7 ~ i , I ~ B , 1 3 ~ , 29si and 119sn NMR
spectra of species generated in solution when trirnethylstannyl or dirnethylphenylsilyl
lithium were reacted with salts of electrophilic aluminum or boron and finally with
1 -2~-oc t~ne .
Results and discussion
The formation of hexabutylditin and tetrabutyltin as side products as well as the
incomplete consumption of l-alkyne even with three equivalents of Bu3SnAlEt2 (90)
and [(BqSn-9-BBN*OMe)-Li+] (94) suggested that stannylalumination and
stannylboronation were competing with other processes. The mechanistic possibilities
shown in Scheme 45, a-h, (shown for Bu3SnAlEt2) illustrate the wide variety of
processes that have literature precedant in related systems.
The reversibility of process a was investigated by conducting the addition of 90
and 91 using an equirnolar ratio of reactants and allowing the reaction to proceed until no
further consumption of allcyne was observed. At this point an equivalent amount of
1-2H-91 was added and the reaction allowed to proceed for a time equivalent to that for
the initial reaction. Protolytic work-up of the reaction followed by analysis of the
vinylstannane adducts by 1H NMR and GC-MS revealed no 2~ incorporation in either
vinyl stannane (92 and 93). These results suggest that the addition of 91 to 90 is not
reversible under these reaction conditions (i.e., 105a and 106a are not converted to 90
and 91).
Process b was shown to be inoperative by conducting the reaction of 90 with
1 2 ~ - 9 1 in the presence of CuCN followed by protolytic workup and examination of the
2~ content of unreacted 91. No dimunition of 2~ in 91 was observed in this
experiment. Thus, Bu3SnAlEt2 (90) did not remove the C1 hydrogen from 91. This
experiment also suggests that the conversion of 105a and 106a to C8H17C=CAlEt2
(107) as shown in process cb is not occurring. To check the forward process ca
reaction of Bu3SnI-I (58) with 107 in the presence of C U + ~ was examined. Recovery of
only ldecyne from this reaction suggests that process ca is not operative.
To investigate if process d occurs, Bu3SnAlEt2 (90) was reacted with Bu3SnH
(58). Without added catalyst no hexabutylditin was observed over 24 h.
Metallometallations are normally complete in 3 h. When either CuCN, CuI,
Pd(PPh3)2C12 or Pd(Ph3P)4 were added to THF solutions of Bu3SnAlEt2 (90)
containing Bu3SnH (58, process da) the only product detected (GC) was
Bu3SnSnBu3. No hexabutylditin was obtained when copper catalysts were reacted with
Bu3SnH alone (process e) but reaction of Bu3SnAlEt2 (90) with CuI and Pd(Ph3P)4
gave hexabutylditin in 69 and 77% yield respectively. Thus, hexabutylditin can be
formed by reaction of Bu3SnAlEt2 with catalyst in the absence of 1-alkyne. These
results are consistent with formation of hexabutylditin via reductive elimination from a
stannyl-alumino-cuprate as shown in process f. The formation of tetrabutyltin from
decomposition of alkali stannides in the presence of hexabutylditin (process g) is a
known process.94
The possiblity that process h is taking place was examined by conducting the
reaction of 107 (prepared from C8H17C&Li and ClAlEt2)with Bu3SnAlEt2 (90) in
the presence of CuCN. No vinyl stannanes were obtained from quenching this reaction
after 24 h. An independent check of reaction h involved reaction of 90 (method b, Table
IV) with 91 in a 1:l molar ratio followed by quenching with 2 ~ ~ ~ . Incorporation of
only one 2~ into each of the vinyl stannane products 92 and 93 (IH NMR) ruled out the
operation of this process.
In the case of Sn-AI reagents the preparative synthetic procedure which was
found to give good yields of products involved addition of the trialkylstannyllithium and
diethylaluminum chloride to the reaction solution prior to cuprous ion. Monitoring these
solutions by 119sn and 1 3 ~ NMR (vide irfra) revealed that reaction between Me3SnLi
and Et2AlC1 occurred prior to addition of the Cu+salt. The initial reagent is formulated as
Me3SnAlEt2 (54), although cuprous ion is undoubtedly an integral pan of the reactive
species. The mechanism (Scheme 46) considered to be operative involves oxidative
addition (a) of R3SnAIFC2 to CuCN to generate a three coordinate Cu(I) species. This is
followed by insertion (b) of the alkyne, a step which does not involve a change in the
oxidation state of copper. This, in turn, is followed by rearrangement (c) to a vinyltin-
cuprate adduct with oxidation of the copper(I) to Cu(III) through addition (d) of a
second equivalent of R3SnAIFQ. The next logical step is reductive elimination (e) of the
vinyl-stannyl-aluminum adduct to regenerate the initial CuO complex, thereby making
the process catalytic.
In the case of CuBr catalyzed stannyl- or silylrnetallations, the initial step involves
transmetallation to yield a stannyl- or silylcopper reagent and metallic electrophile as
shown in scheme 47 (step a). Silyl- (38) and stannylcopper (49) reagents are highly
reactive toward 1-alkynes due to the presence of an empty coordination site and the
partial positive charge on copper (step b). Species polarized as in 38 or 49 would be
expected to react with 1-alkynes to produce vinyl-silyl or stannylcopper adducts in which
the silyl-
R3SnA1R', = Bu3SnAlEt2 R3SnA1Re2 = Mc3SnA1Et2 S = Solvent
6. 6- 1 ; R 1 ' - E CI-I
1 I
6-! ! 6' AIR', R3Sn - --Cu I +
or stmnyl- group is attached to the more substituted center of the allcyne (step c).
Transmetallation of the vinyl-copper bond by metallic electrophile (step d) generated in
133
situ followed by reductive elimination (step e) would result in the eventual production of
2-metallated- 1 -alkenes (Scheme 47).
R3Sn-M(R), + CuBr
R3Sn = Bu3Sn R3Sn = M ~ s n M(R), = AIEtz M(R), = BBN S = Solvent
Stannylalumination also proceeds well when diethylalurninum chloride is added
to the reaction mixture after the triakylstannate anion, Cu+ salt and alkyne (Scheme 48).
In this case we consider-the reaction to proceed via initial formation of the LO
triakylstannylcyanocuprate (48, a). Complexation of this species with akyne (b)
followed by rearrangement (c) to a vinyltincuprate adduct yields a Cu(1) species.
R3Sn = Bu3Sn R,Sn = Mc3Sn S = Solvent
Reaction of this complex with diethylaluminum chloride (d) would give a Cu(m)
species with an expected propensity for reductive elimination (e) to the catalytic Cu+
species and the ~in~l-sta&~l-aluxninum adduct in which the silicon or the tin moiety is at
more substituted end of alkyne. Indeed, quenching the vinylstannyl-copper or the
vinylsilyl-copper adduct generated in the reaction of "R3MCuU (M = Sn or Si, i.e., 26,
38 or 48,492 with 1-nonyne by either MeOH, Et2AlC1, Br-9-BBN led to the exclusive
formation of 2-stannyl- or 2-silyated alkenes. A rationale for the observed mode of
addition is provided by the electronic effects governing the transition state as shown in
Schemes 47 and 48 (step c). Quantum mechanical and molecular orbital calculations
have shown that the electron density distribution in case of l-propyne95 to be in the
direction that one expects an anion to attack at the central carbon atom resulting in the
formation of 2-metallated alkenes.
It can be envisioned that steric environment around the copper moiety can also
dictate the regiochemistry in silyl- and stannylcuprations of 1-alkynes. The latter
consideration is important was suggested by the observation that addition of highly
coordinating solvent like HMPA to R3Si(Cu)(CN)Li (26) resulted in the formation of the
reverse regioisomer (i.e., 1-silyated alkene). That LO cuprates possess a sterically
demanding Cu center has been suggested from our NMR studies on (26), R3SiCu (38),
RjSn(Cu)(CN)Li (48) and R3SnCu (49). The 1 3 ~ 2% and 1 1 9 ~ n NMR spectra of
these reagents consists of several broad lines indicative of oligomeric species in solution.
In highly coordinating solvents like HMPA these cuprates tend to be less aggregated as
seen by appearance of sharp signals in the 1 3 ~ NMR spectra of 26 (Figure 18).
According to this ?mposal the change in regioselectivity on the addition of HMPA to 26
and 1-octyne (Scheme 12) was due to &aggregation of this reagent in this solvent.
Further support of this hypothesis comes from the exclusive formation of 1-silyated
alkenes on the addition of (PhMe2Si)2Cu(CN)Li2 (28) to 1-alkynes which was shown
alkenes on the addition of (PhMe2Si)2Cu(CN)Li2 (28) to 1-alkynes which was shown
to be less aggregated in THF. Thus, it is envisioned that the additions of mixed stannyl-
(R3Sn)Cu(R')(CN)Li2 -(56) or mixed silylcuprates (R3Si)Cu(R')(CN)Li2 (43) can
result in the formation of either regioisomer depending upon the steric bulk of R'..
Regioselectivity in metallometallations is influenced by the catalyst employed as
well as by the metals and steric bulk of the alkyl groups in the bismetalloid. According to
Scheme 46, a change in regiochemistry in the cuprous ion catalyzed silyl- or
stannylmetallations of 1-alkynes is due to a change in the regioselectivity in step c. The
differences in electronegativities of the metalloid partners would be expected to give the
bond connecting them some polar character and the direction of this polarity would be
expected to be dictated by the nature of appended ligands (denoted 2 ~ ) . The metalloids
known to participate in stannylmetallations are B, Si, Cu, Zn, Al, Zr, and Mg (written in
order of decreasing electronegativity). In the case of copper catalyzed
stannylmetallations, tin should be more electronegative than copper associated with the
attached metal (2M) in those cases where the latter is strongly electronegative. In cases
where 2~ is weakly elemnegative it is possible that the copper would be more electron
rich and the polarity of the tin-copper bond could reverse (Scheme 49).
According to this proposal, it is envisioned that the metal partners of tin with low
electronegativity (Mg, Zr, Al) will be better donors to the copper center in the proposed
addition complex (Step c, Scheme 46) and that this will enhance addition in the reverse
mode (Mg > Zr > Al). For those metals less electronegative than copper this trend
certainly seems evident. Additions of stannylaluminum reagents give noxmal
regiochernistry while addition of stannylrnagnesium reagents proceed with reverse
regiochemistry.70 It is expected that at low temperature Sn-Zr will give good
regioslectivity in favor of reverse addition. Also, bulky74 ligands on ZM should result in
the opposite regioisomer 93a. On the other hand, metalloids with high electronegativity
(Zn, Si, B) will act to enhance addition in the normal mode (B > Si > Zn). The Sn-
Zn,70~n-~i73 and Sn-~72 additions are reported to give excellent regioselectivity in
favor of the normal addition product 92a (i.e., 2-stannyl-1-alkenes).
Reversal of regioselectivity of stannylaluminations in the presence of HMPA
would, in this proposal, be attributable to its ability to efficiently complex 2 ~ , thus,
accentuating the negative chwe on copper and favoring generation of the regioisomer
93a (1-stannyl-1-alkenes). 93a can also result from the addition of less aggregated
trialkylstannylcopper species to 1-allyes.
Thus, it is possible to envision a change in regiochemistry for stannylalumination
in the presence of HMPA or bulky ligands without ~quiring either stannyl- or
silylcupration or subsequent steps to be reversible.
Scheme 49
Spectroscopic studies
Metallometallations are usually conducted by addition of 1-allye to the
preformed bimetallic reagent, followed by addition of cuprous salt. This sequence of
reagent addition prevents formation of h e r by-products which can arise if the cuprous
salt is added to the bimetallic reagents prior to akyne. Since stannyl- or silylmetalloids do
not react with alkynes without the &(I) catalyst, it was envisioned that the Sn-Cu-
metalloid sequence of addition would generate the same species as those involving the
Sn-metalloid-& sequence of addition. Thus, both sequences would result in the in situ
generation of stannyl- or silylcuprates from stannyl- or silylmetalloids as shown in
Scheme 3 1.
The 1 3 ~ NMR spectra of solutions resulting from the reaction of Et2AlCl with
Me3SnLi (47) at -70•‹C in THF exhibited a singlet at -7.5 pprn due to the methyl carbons
bound to tin in Me3SnAlEt2 (54) alongwith a singlet for MeqSn at -9.3 pprn (Me3SnLi
was prepared from the cleavage of Me6Sn2 with MeLi Figure 37a). The 1 1 9 ~ n NMR
spectra of this solution revealed a broad signal for Me$h41Et2 (54) at -126 pprn with
respect to MeqSn (0.0 ppm, an internal reference). This signal broadening may be
attributed to the coupling of tin to quadrupolar aluminum. The 7 ~ i NMR exhibited a
sharp signal at 0.0 pprn implying the formation of LiCl in these solutions.
Addition of a THF solution of CuCN*2LiC1(1.0 equivalent) to 54 resulted in the
formation of a new species exhibiting a broad 1 3 ~ signal at -4.5 pprn (Figure 37b). The
chemical shift of this signal corresponded to that obtained earlier for the solutions
generated by mixture of equimolar amounts of Me3SnLi (47) and CuCN.2LiCl (Figure
33b). More remarkable was the obsevation that admixture of equimolar ratios of
MesSnCu(CN)Li (48) and Et2AlC1 (Figure 37c) resulted in the same chemical shifts as
obtained earlier (Figure 37b). Similarly, the 1 1 9 ~ n NMR chemical shifts from both the
reagent combinations [i.e., Me3SnCu(CN)Li (48,6 -150 ppm)-Et2AlCl and
Me3SnAlEt2 (54,6 -126)-CuCNa2LiCl were identical indicating that same species
(Me3SnCu(CN)AlEt2, 6 -156, step a in Scheme 461 is obtained when the ratio of
stannyl anion to copper cation and aluminum cation is 1 : 1 : 1.
Me, Sn AI Et -7.5
I I3c -70•‹C
PPM
0 -1 0 -20 PPM
0 -1 0 -20 PPM
Figure 3 7 . 1 3 ~ NMR spectra of (a) Me3SnAIEt2 (b) MegSnAlEt2 + CuCN (c)
Me3SnCu(CN)Li + Et2AK1; the spectra were run at -70•‹C.
Addition of 1-2~-octyne to the clear solution of Me3snAlE~ (54) followed by
CuCN (step b, Scheme 46) resulted in the formation of a wine red solution that exhibited
a l 1 9 ~ n NMR signal at -50.0 pprn This signal is very close to the one obtained for the
vinyl-tin (6 -53.0) in species 79a (Scheme 38, where M(R)n = Li, vide ir3fra). The 2~
NMR exhibited a broad signal at 4.1 pprn indicating the formation of a vinylic carbon.
Addition of a further equivalent of Me3SnAlEt2 (54, step d, Scheme 46) resulted in the
appearance of a new vinyl 2H peak at 7.2 pprn which we attribute to the vinytin-
aluminum adduct, 79b (step e, Scheme 46).
We next focussed attention on addition of stannylcuprate (48) to 1-alkynes. The
1 1 9 ~ n NMR spectra of solutions generated from the addition of one equivalent of
Me3SnCu(CN)Li (48,6 -15 1, step a, Scheme 48) to 1 -2~-octyne (step b, Scheme 48)
at -50•‹C resulted in the appearance of signals for a vinyltin-copper adduct at -53.0 ppm.
The 2~ signal of the 12~-octyne (1.45 ppm) shifted to 4.1 ppm. Addition of one
equivalent of Et2AlCl (step d, Scheme 48) resulted in a downfield shift of the 2H signal
to 7.1 pprn (Figure 38). It should be recalled that this signal has the same chemical shift
as that obtained by the addition of 54 to 1-ZH-octyne followed by CuCN (vide supra).
These NMR results cogently attest to the involvement of a vinylstannylcopper species
such as 79a that are converted to a species formulated as 79b (Scheme 38 or step e,
Schemes 46 and 48).
The unusual upfield shift of the viny12H (vinyl protons generally absorb around
6-7 ppm) obtained for vinyltin-copper species (79a, Scheme 38 and also shown in step
c, Schemes 46 and 48) can be understood in terms of changes in the x electron density at
carbon due to metal-akyne coordination. According to the Dewar-Chatt-Duncanson
model26 metal coordination to an alkyne should result in net shielding of the vinyl
carbon 1 3 ~ and 1~ chemical shifts as a result of perturbation due to x-backbonding.97
This may be attributed to an increase in electron density. On the other hand, perturbation
due to a-bonding should result in net deshielding of the chemical shifts of the attendent
carbon and proton due to a demase in electron density. Appearance of the viny1JH
signal at around 4 ppm during addition of stannylcopper (48 and 49) and silylcopper
(26,27 and 28) to 1-akynes (similar upfield shifts at 6 3.3 were obtained) suggests that
n-back bonding is more dominant in the akyne complexes of cur. This may be
attributed to the full dl0 electron configuration of Cum complexes which favours the
transfer of electron density from the filled d orbitals of copper to the empty x*-orbitals of
the akyne. Addition of metallic electrophiles (e.g., Et2AlCl or Br-9-BBN) should result
8.0 6.0 4.0 2.0 0
PPM
Figure 3 8 . 2 ~ NMR spectra of Me3SnCu(CN)Li + 1-2H-octyne + Et2AICI; the spectra
were run at -70•‹C.
in the downfield shift of the vinyI-carbon as well as vinyl-2~ due to transfer of electron
density to the electron deficient metal.
Copper(I) catalyifed stannylboronation of 1-alkynes by low-tempem NMR
spectroscopy (Scheme 50) was next studied. Reaction of equirnolar ratios of Me3SnLi
(47,6 1%n, -187.7) and 9-BBNaOMe (6 I ~ B , 55.4) resulted in the formation of
(Me3Sn-g-BBN*OMe)-Li+ (55). This species exhibited a broad 119~n NMR signal at
-109.5 ppm at -50•‹C due to the quadrupolar coupling caused by boron (log, natural
abundance: 19.5846, spin: 3; I ~ B , natural abundance: 80.42, spin: 3/2).98 Cooling the
solution to -8S•‹C resulted in considerable sharpening of the signal. The IB spectrum
exhibited a broad signal at 2.3 ppm. This value is repoxted for boron compounds bound
to four ligands. Addition of 12~-octyne to the preformed reagent,
[(Me3Sn-9-BBN*OMe)-Li+] (55) followed by CuBpMe2S resulted in the formation of
the vinyltin-boron adduct as indicated by the appearance of vinYltin'(ll%n, 6 -50.7) and
viny12H NMR (5.7 ppm) signals attributable to this species.
The 119sn NMR spectm obtained from the addition of 1-2~-octyne to
Me3SnCu (49, Figure 40% step b, Scheme 47) exhibited a singlet at -59.0 ppm (figure.
40b). This signal is attributed to the vinyltin-copper adduct, 79a (Scheme 38, where
M(R)n = Li, step c, Scheme 47). The 2~ NMR exhibited a broad signal at 6.1 ppm
indicating the formation of a vinylic species (Figure 41% Scheme 50). Remarkably,
addition of an equivalent of Br-9-BBN (6, I ~ B , 84.0, step d, Scheme 49) followed by
an equivalent of NaOMe resulted in the formation of the same intermediate which was
obtained earlier for Cum catalyzed stannylboration of 1-2~-octyne (Scheme 50).
I I I I I I L 0 -50 -150 -250
PPM
u u 0 -50 -1 00 -1 50
PPM
Figure 39. 11%n NMR of (a) Me3SnCu (b) MegSnCu + ~ - ~ ~ - o c t ~ n e ; the spectra
were run at -50•‹C.
Mc,SnCu:Oc tyne-d
1 PPM
Mc,SnCu:Oclync-d:DrDBN 1:l:l
I I I 1 L
8 .O 6 .O 4.0. 2.0 0 PPM
Figure 40. 2~ NMR of (a) Me3SnCu + 1-2~ -0c t~ne (b) Me3SnCu + 1-2~ -oc tpe + Br-9-BBN + NaOMe; the spectra were run at -70•‹C.
Conclusion
Thus, in conclusion copper (I) catalyzed metallometallation of 1-alkynes can be
visualized as initial metallocuprations of 1-alkynes followed by transmetallation of the
vinylmetallocopper adduct by metallic electrophile generated in situ in the reaction
mixture. These reactions are compatible with polar functional groups. In situ addition of
a proton source is not necessary to achieve high conversion of akyne as the 12-
dimetallo-adducts generated can be easily captured by electrophiles.
EXPERIMENTAL
All glassware and syringes were dried in an oven overnight at 120•‹C, and
glassware was flame dried under vacuum and flushed with argon immediately prior to
use. Syringes were flushed with argon and kept under positive argon pressure while
cooling until use. Transfer of reagents was performed by with syringes equipped with
stainless steel needles. Reactions were carried out in three-necked round bottom flasks
equipped with filtration units and teflon-coated magnetic stirring bars. Usual workup
involved quenching of the reaction with 1N HCI, extraction of the organic layer with
Et20 (2 x 5 mL), back-washing the combined organic extracts with satd. NHqCl(2 x 5
mL) and drying of the organic layer over anhydrous MgSO4.
Transfer of CuCN and CuBpMe2S took place in a glove bag. CuBr*Me2S was
purified by the method of ~ouse.99a All alkyllithiums were freshly titrated before
99b use.
Tetrahydrofuran was freshly distilled over potassium benzophenone-ketyl.
Hexamethylphosphoramide and diisopropylamine were distilled over calcium hydride
and stored over activated 3 A molecular sieves. Unless otherwise stated, other chemicals
obtained from commercial sources were used without further purification.
Low-temperature 19sn NMR experiments were conducted on a Bruker WM-400
spectrometer with an operating frequency of 149.197 MHz. A typical set of parameters
utilized a spectral width of 50000 Hz, 8K of memory, 11 Hzfdata point, an acquisition
time of 0.09 s and a 55' pulse of 35 ps. The decoupler was turned on during acquisition
and off during the relaxation delay (4 s) in order to supress the negative nOe of 1 %n. A
line broadening of 20 Hz was applied to all spectra. Spectra were recorded in THF that
contained M a n as internal reference.
LOW-temperat~re-~~~i NMR experiments were conducted on a Brdcer WM-400
spectrometer with an operating frequency of 79.495 MHz. A typical set of parameters
utilized a spectral width of 20000 Hz, 8K of memory, 2.44 Wdata point, an acquisition
time of 0.204 s and a 15' pulse of 10 ps. The decoupler was turned on during acquisition
and off during the relaxation delay (4 s) in order to supress the negative nOe of 29~i . A
line broadening of 20 Hz was applied to all spectra. Spectra were recorded in THF that
contained Me& as internal reference.
l3C NMR spectra (> -30•‹C) were obtained on Bruker WM-400 spectrometer at a
frequency of 100.61 MHz. Parameters for the spectral acquisition typically involved
a spectral width of 22000 Hz, 32K of memory, 1.32 Wdata point, an acquisition time
of 0.75 s and a 13.5' pulse of 9 ps. The spectra were recorded in THF solutions unless
otherwise specified and were referenced to THF, a = 26.0 ppm, P = 68.2 ppm. Inverse-
gated decoupling was employed.
l3C NMR spectra were also obtained on Varian XL3O spectrometer (run at
tempcn~rnes c -30•‹C ) with an operating frequency of 75.46 MHz. Parameters for the 1 3 ~
spectral acquisition typically involved a spectral width of 15000 Hz, 32K of memory, an
acquisition time of 0.4 s and a 60' pulse of 12 ps. The spectra were recorded in THF
solutions unless otherwise specified and were referenced to THF, a = 25.3 ppm, P =
67.41 ppm.
Low-temperature 7Li NMR spectra were obtained on a Varian XL-300
spectrometer with an operating frequency of 116.6 MHz with a sweep width of 20000 Hz,
16K of memory, 1.25 Hddata point, an acquisition time of 0.4 s and a 55' pulse of 12 ps.
A line broadening of 10 Hz was applied to all spectra. 7 ~ i chemical shifts were referenced
with respect to 0.5M LiWD30D (6 0 ppm) in a capillary insert.
Low-temperature JH NMR spectra were recorded on a Varian XI.,-300
spectromter in THFdg and were referenced to MeqSn (6 0). Low-temperature IB NMR
spectra were recorded on a Bruker WM-400 spectrometer with BF3eEt20 (6 0) as internal
standard.
A vacuum-jacketed glass dewar measuring 7.5 x 16.0 cm (id 5.5 cm) was designed
with tapering bottom to fit in the cup of the vortex mixer. All NMR samples were stirred
while cooling at the indicated temperatures in this dewar.
1~ NMR and 119~n spectra of isolated products were recorded on a Bruker
WM-400 spectrometer in CDCl3 with CHCl3 (6 7.25) and MeqSn (6 0) respectively, as
internal standards. The tin-proton coupling constants (Jsn-H) are given as an average of
the 1 17sn and 1 19sn values.
Gas chromatbgraphic analyses utilized a Hewlett-Packard 5880A instrument
equipped with a with a flame ionization detector and employing a JW fused silica DB-1
capillary column (15 m x 0.25 mm), with a linear temperature gradient. The purity of all
the title compounds was >95% as judged by gas chromatographic analysis with
dodecane as an internal standard. Chromatographic purifications were carried out with
E.M. Merck silica gel (60, particle size 0.040-0.063 mm). Low resolution mass spectra
were obtained with an HP 5985B GC-MS system with electron-impact ionization at 70
ev wme me mgn resolution mass spectra were obtained on a Kratos MS50 RFA mass
spectromter (University of British Columbia, regional facility). For compounds
containing a BugSn group, molecular mass measurements are based on l2O~n and use
the (M+- Bu) peak. Infrared (IR) spectra were recorded in THF solutions with Perkin-
Elmer Model 283 spectrophotometer.
PREPARATION OF TRIALKYLSILYLCUPRATES
2% NMR Sample Preparations
Preparation of PhMe2SiLi (25) in THF: Dimethylphenylsilyl chloride
(3.6 g, 21.0 mmol) was stirred with small pieces of lithium (0.450 g, 64.0 mmol) in
THF (20 mL) at -5OC in an icelsalt bath.33a Reaction was initiated by immersion of the
reaction flask in a sonicator for 30 min and then stirred overnight at -5OC.
DimethylphenyIsilyllithium was titrated according to the procedun of Fleming et d27b
Preparation of LiCl-Free PhMe2SiLi (25) in THF: According to the
procedure of ~ilman,33a to a solution of 12-diphenyl- 1,122- tetramethyldisilane
(prepared according to the procedure of ~ilman33b) (4.6 g, 17.0 mmol) in THF (10 mL)
were added small pieces of lithium (24 mg, 34.0 mmol). The reaction was initiated in a
sonicator bath for 30 min and then stirred overnight at -5OC. The resultant green solution
gave a negative halogen test.
Preparation of PhMe2SiLi (25) in DMS: 1,2-Diphenyl- 1, l,2,2-
tewmethyldisilane33b (3.6 g, 21.0 mmol) was stirred with small pieces of lithium
(0.450 g, 64.0 mmol) h THF (20 mL) at - 5OC in an icelsalt bath. The reaction was
initiated by immersion of the reaction flask in a sonicator for 30 min and then stirred
overnight at -5OC. THF was removed under vacuum and replaced with an equal volume
of DMS. This procedure was repeated three times. Dirnethylphenylsilyllithium was
titrated according to the procedure of Fleming et al.2'b
Generation of Silylcuprates from CuCN
Preparation of PhMe2SiCu(CN)Li (26): CuCN (0.18 g, 2.0 -01) was
placed in a 10 mm NMR tube, equipped with an argon inlet. The tube was repeatedly
(3x) evacuated and purged with argon. Me& (0.5 mL) was injected, the reaction was
cooled to -50•‹C and dimethylphenylsilyIlithium in THF (1.8 mL, 2.0 mrnol) added
dropwise. The solution was stirred on a vortex mixer at -50•‹C in a custom built dewar for
20 min. The NMR spectrum was then immediately recprded.
Preparation of (PhMe2Si)zCu(CN)Li2 (28): A THF solution of
PhMe2SiLi (25, 1.8 mL, 2.0 -01) was added to a 10 mrn NMR tube containing a THF
solution of PhMe2SiCu(CN)Li (26, vide supra) at -50•‹C. The deep red solution was
stirred for 20 min at -50•‹C prior to examination by NMR and IR spectroscopy.
Preparation of (PhMe2Si)2CuLi2 (29): To a THF solution of
(PhMe2Si)2Cu(CN)Li2 (28,2.0 mmol) prepared as outlined above was added a THF
solution of PhMe2SiLi (25, 1.8 mL, 2.0 mmol) at -50•‹C. The reaction was stirred for 20
min at -50•‹C before examination by NMR and IR spectroscopy.
Preparation of (PhMezSi)3CuLi2 (29) with excess PhMe2SiLi (25):
To a THF solution of (PhMe2Si)3CuLi2 (29,2.0 mmol) prepared as outlined above was
added a THF solution of PhMe2SiLi (25,l.g mL, 2.0 mmol) at -50•‹C. The reaction was
stirred for 20 min at -50•‹C before examination by NMR spectroscopy.
Regeneration of (PhMe2Si)~Cu (CN)Li2, 28 from 29: CuCN (0.09 g,
1.0 mmol) was added to the solution of 29 (2.0 mmol). The reaction was stirred for 20
min at -50•‹C before e x h a t i o n by NMR spectroscopy.
Regeneration of PhMezSiCu(CN)Li (26) from 28: CuCN (0.18 g, 2.0
rnmol) was added to the NMR tube containing silylcyanocuprate 28. The reaction was
stirred for 20 min at -50•‹C and then examined by NMR spectroscopy.
Generation of Silylcuprates from CuBreMe2S
Preparation of (PhMe2Si)Cu (38): CuBr*Me2S (0.41 g, 2.0 mmol) was
placed in a 10 mm NMR tube, equipped with an argon inlet. The tube was repeatedly
(3x) evacuated and purged with argon. Me& (0.5 mL) was injected, the solution was
cooled to -50•‹C and dimethylphenysilyllithium in THF (25, 1.8 mL, 2.0 mmol) added
dropwise. The reaction mixture was stirred on a vortex mixer at -50•‹C in the custom built
dewar for 20 min. The NMR spectrum was then immediately recorded.
Preparation of (PhMe2Si)~CuLi (27): A THF solution of PhMe2SiLi (1.8
mL, 2.0 mmol) was added to a 10 mm NMR tube containing a THF solution of
PhMe2SiCu (38,2.0 mrnol, vide supra) at -50•‹C. The resultant deep redsolution was
stirred for 20 min at -50•‹C prior to examination by NMR spectroscopy.
Preparation of (PhMe~Si)3CuLi2 (29): To a THF solution of
(PhMe2Si)2CuLi (2.0 mmol) prepared as outlined above, was added a THF solution of
PhMe2SiLi (1.8 mL, 2.0 mmol) at -50•‹C. The reaction mixture was stirred for 20 min at
-50•‹C before examination by NMR spectroscopy.
Preparation of (PhMe2Si)3CuLi2 (29) with excess PhMe2SiLi (25):
To a THF solution of (PhMe2Si)3CuLi2 (29,2.0 mmol), prepared as above from
CUBPM~~S, was added a THF solution of PhMe2SiLi (1.8 mL, 2.0 mmol) at -50•‹C.
The reaction mixture was stirred for 20 min at -50•‹C before examination by NMR
spectroscopy.
Regeneration of (PhMe2Si)2CuLi (27) from 29 prepared from
CuBr.Me2S: CuBr.Me2S (0.205 g, 1.0 mmol) was added to 2.0 mmol of the solution
generated by mixing PhMe2SiLi (25) and CuBr.Me2S in a 3:l ratio. The reaction
mixture was stirred for 20 min at -50•‹C before examination by NMR spectroscopy.
Regeneration of PhMe2SiCu (38) from 27: CuBreMe2S (0.41 g, 2.0
mmol) was added to the NMR tube containing silylcuprate (27,2.0 mmol). The reaction
was stirred for 20 min at -50•‹C and then examined by NMR spectroscopy.
Generation of Mixed Silylcuprates
Preparation of PhMe~Si(Me)cu(cN)Li~ (43) by Reaction of
PhMe2SiLi (25) and MeCu(CN)Li (16): CuCN (0.18 g, 2.0 mmol) was added to
a 10 mm NMR tube, equipped with an argon inlet. The solution was cooled to -50•‹C and
MeLi (1.4 mL, 2.0 mmol) in Et20 was added via a syringe. The reaction mixture was
stirred on the vortex mixer in a custom built dewar for 10 min.
Dimethylphenylsilyllithium (25) in THF (1.8 mL, 2.0 mmol) was added dropwise to this
clear solution at -50•‹C. The reaction turned deep red in color. The NMR spectrum was
recorded immediately.
Preparation of 43 by Reaction of PhMe2SiCu(CN)Li (26) and
MeLi: MeLi (1.4 mL, 2.0 mmol) was added to a solution of 26 (2.0 mmol, vide supra)
at -50•‹C. .The reaction mixture was stirred on a vortex mixer at -50•‹C in a custom built
&war for 20 min. The NMR spectrum was immediately recorded.
Preparation of 43 by Reaction of (PhMe2Si)2Cu(CN)Li2 (28) and
MezCu(CN)Liz (17): CuCN (0.09 g, 1.0 mmol) was added to a flask, equipped with
an argon inlet. The solution was cooled to -50•‹C where MeLi (1.4 mL, 2.0 mmol) was
introduced dropwise to generate a clear solution of 17. After 10 min this solution was
t ransfed via a pre-cooled cannula into cuprate 28 (2.0 mmol, vide supra) which was
also maintained at below -50•‹C. The resulting deep red solution was stirred for 20 min
before recording the NMR spectrum
Preparation of (PhMe2Si)2(Me)CuLi2 (44) by Reaction of
PhMezSiLi (25) and MeCu(CN)Li (16): CuCN (0.09 g, 1.0 mmol) was added to
a 10 mm NMR tube, equipped with an argon inlet. The solution was cooled to -50•‹C and
MeLi (0.7 mL, 1.0 mrnol) in Et20 was added via a syringe. The reaction mixture was
stirred on a vortex mixer in a custom built dewar for 10 min. Dimethylphenylsilyllithium
(25) in THF (1.8 mL, 2.0 mmol) was added dropwise to this clear solution at
-78OC. The reaction turned deep red in color. The NMR spectnun was recorded
immediately.
Preparation of 44 from Reaction of (PhMe2Si)2Cu(CN)Li2 (28)
and MeLi: MeLi in Et20 (1.4 mL, 2.0 mrnol) was introduced dropwise via a syringe
Preparation of 44 from Reaction of (PhMe2Si)2Cu (CN)Li2 (28)
and MeLi: MeLi in Et20 (1.4 mL, 2.0 mmol) was introduced dropwise via a syringe
to a solution of 28 (2.0 mmol) at -78OC. The resulting deep red solution was stirred for
20 min before recording the NMR spectrum.
Preparation of PhMe2Si(Me)2CuLi2 (45) by Reaction of
PhMe2SiLi (25) MeLi and CuCN: CuCN (0.09 g, 1.0 mmol) was added to a 10
mm NMR tube, equipped with an argon inlet. The solution was cooled to -50•‹C and
MeLi (0.7 mL, 1.0 mmol) in EQO was added via a syringe followed by
dimethylphenylsilyllithium (25) in THF (2.7 mL, 3.0 mmol) at -78OC. The reaction
mixture was stirred on a vortex mixer in a custom built dewar for 10 min. The reaction
turned deep red in color. The NMR spectrum was recorded immediately.
Preparation of 45 by Reaction of (PhMe2Si)3CuLi2 (29) and MeLi:
MeLi in EQO (0.7 mL, 1.0 mmol) was introduced dropwise via a syringe to a solution
of 29 (1.0 rnmol, vide supra) at -78OC. The resulting deep red solution was stirred for 20
min before recording the NMR spectrum.
1 3 ~ and 1~ NMR Sample Preparation
Preparation of CuCN.2LiCl: THF (1 1.0 rnL) was added to a mixture of
CuCN (0.98 g, 11.0 mmol) and LiCl(0.95 g, 22.0 mmol) in a round bottomed flask
under argon. A clear faint yellow solution was obtained after 0.5 h of stirring. This
solution was used as the Cu(1)CN source for all the l 3 ~ and IH NMR sample
preparations unless otherwise specified.
Preparation of CuBr.2LiCl: THF (1 1.0 rnL) was added to a mixture of
CuBr.Me2S (2.25 g, 11.0 mmol) and LiCl(0.95 g, 22.0 mmol) in a round bottomed
flask under argon. A clear dark yellow solution was obtained after 0.5 h of stirring. This
solution was used as the CuOBr source for all the 1% and 1~ NMR sample
preparations.
Generation of Silylcuprates from CuCN
Preparation of 26: The above THF solution of CuCN (0.5 rnL, 0.5 mmol)
was added to a 5 mm NMR tube, equipped with an argon inlet. The solution was cooled
to -78OC and dimethylphenylsilyllithium in THF (0.6 mL, 0.5 mmol) added dropwise.
The reaction mixture was stirred on a vortex mixer in a custom built dewar for 20 min.
The spectra were recorded immediately.
Preparation of 26 in THFIHMPA: A solution of CuCN in THF (0.5 mL,
0.5 mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The solution
was cooled to -78OC and dimethylphenylsilyllithium in THF (0.6 mL, 0.5 mrnol) added
dropwise followed by addition of HMPA (0.5 mL) via a syringe. The reaction was
stirred on a vortex mixer in a custom built dewar for 20 rnin. The spectra were recorded
immediately.
Preparation of 28: The above THF solution of CuCN (0.25 mL, 0.25 mmol)
was added to a 5 mm NMR tube, equipped with an argon inlet. The solution was cooled
to -78OC and dimethylphenylsilyllithium in THF (0.6 mL, 0.5 mrnol) added dropwise.
The reaction mixture was stirred on a vortex mixer in a custom built dewar for 20 min.
The spectra were recorded immediately.
Preparation of 29: The above THF solution of CuCN (0.25 mL, 0.25 mmol)
was added to a 5 mm NMR tube, equipped with an argon inlet. The solution was cooled
to -78OC and dimethylphenylsilyllithium in THF (0.9 mL, 0.75 mmol) added dropwise.
The reaction was stirred on a vortex mixer in a custom built dewar for 20 min. The
spectra were recorded immediately.
Preparation of 29 with excess 25: The above THF solution of CuCN
(0.125 mL, 0.125 mmol) was added to a 5 mm NMR tube, equipped with an argon inlet.
The solution was cooled to -78OC and dimethylphenylsilyllithium in THF (0.6 mL, 0.5
mmol) added dropwise. The reaction was stirred on a vortex mixer in a custom built
dewar for 20 min. The spectra were recorded immediately.
Preparation of 26 for 1~ NMR analysis: The above THF solution of
CuCN (0.5 mL, 0.5 mmol) was added to a 5 mm NMR tube, equipped with an argon
inlet. The solution was cooled to -78OC and di&thylphenylsilyllithium in THF-d8 (0.6
mL, 0.5 mmol) added dropwise. The reaction mixture was stirred on a vortex mixer in a
custom built dewar for 20 min. The spectra were recorded immediately.
Preparation of 28 and 29 for 1~ NMR analysis: These solutions were
prepared in THF-d8 as described above.
158
Generation of LiC1-free Silylcuprates from CuCN
Preparation of PhMe2SiCu(CN)Li (26): CuCN (0.06 g, 0.66 mmol) was
placed in a 5 mm NMR tube, equipped with an argon inlet and a capillary insert
containing 1M solution of LiCl in M ~ O ~ H . The tube was repeatedly (3x) evacuated and
purged with argon. LiC1-free dirnethylphenysilyllithium in THF (0.6 mL, 0.66 mmol)
added dropwise at -70•‹C. The solution was stirred on a vortex mixer at -70•‹C in a custom
built dewar for 20 min. The NMR spectrum was then immediately recorded.
Preparation of 28 and 29: These were prepared precisely as described
above. Both the solutions were green in color. Addition of LiCl resulted in the familiar
red silylcuprates.
Generation of Silylcuprates from CuBr*Me2S
Preparation of PhMe2SiCu (38): CuBr*Me2S (0.5 mL, 0.5 mmol) was
added to a 5 mm NMR tube, equipped with an argon inlet. The reaction was cooled to
-78OC and dimethylphenylsilyUithium in THF (0.6 mL, 0.5 mmol) added dropwise. The
solution was stirred on a vortex mixer in a custom built dewar for 20 rnin. The slurry
was used as such.
Preparation of (PhMe2Si)2CuLi (27): This cuprate was generated as
described above. Thus, CuBr*MeaS (0.25 mL, 0.25 mmol) was added to a 5 mm NMR
tube, equipped with an argon inlet. The reaction was cooled to -78OC and
dimethylphenylsilyUithium in THF (0.6 mL, 0.5 rnmol) added dropwise. The solution
was stirred on a vortex mixer in a custom built dewar for 20 min before examination of
the NMR spectrum.
Preparation of (PhMe2Si)3CuLi2 (29): CuBr.Me2S (0.125 rnL, 0.125
mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The reaction was
cooled to -78OC and dimethylphenylsilyllithium in THF (0.9 mL, 0.75 mmol) added
dropwise. The solution was stirred on a vortex mixer in a custom built dewar for 20 min
before examination of the NMR spectrum.
Generation of Silylcuprates from CuBr in DMS
Preparation of (PhMe2Si)2CuLi (27) in DMS: CuBreMe2S (0.05 g 0.25
mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The reaction was
cooled to -78OC and dimethylphenylsilyIlithium in DMS (0.45 mL, 0.5 mrnol) added
dropwise. The solution was stirred on a vortex mixer (vide supra). The spectra were
recorded at -85OC immediately. In a separate set of experiments, solutions of silylcuprate
27 in DMS were prepared in a flask equippped with a filtration unit. The silylcuprate was
filtered under argon via a cannula to the pre-cooled 5 rnm NMR tube. Both the solutions
exhibited same resonances in the 13c NMR spectra. For the 13c analysis the samples
were prepared without filtration whereas the 7 ~ i NMR samples were filtered prior to
examination by NMR spectroscopy.
Preparation of LiC1-free (PhMezSi)2CuLi (27) in THF: CuBr.Me2S
(0.05 g 0.25 mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The
reaction was cooled to -78OC and LiC1-free dimethylphenylsilyllithium in THF (0.45 mL,
0.5 mmol) added dropwise. The solution was stirred on a vortex mixer. The spectra were
recorded at -85OC immediately.
Preparation of (PhMe2Si)2CuLi*LiI (39) in DMS: (PhMe2Si)2CuLi*LiI
was prepared as above except for the substitution of CuI (0.047 g 0.25 mmol) for
CuBrMe2S.
Preparation of (PhMe2Si)3CuLi2 (29) in DMS:
DimethylphenylsilyIlithium in DMS (0.45 mL, 0.5 mmol) was added to the silylcuprate
27 (0.5 mrnol) at -78OC. The solution was stirred on a vortex mixer (vide supra). The
spectra were recorded at -8S•‹C immediately.
Generation of Mixed Silylcuprates
Preparation of 43 by Reaction of 25 and 16: CuCN (0.25 mL, 0.25
mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The solution was
cooled to -50•‹C and MeLi (0.18 mL, 0.25 mmol) in Et20 was added via a syringe. The
reaction mixture was stirred on a vortex mixer in a custom built dewar for 10 min. Upon
addition of dimethylphenylsilyllithium in THF (0.3 mL, 0.25 mmol) dropwise to this
clear solution at -78OC the reaction turned deep red. The NMR spectrum was recorded
immediately.
Preparation of 43 by Reaction of 26 and MeLi: CuCN (0.25 mL, 0.25
mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The solution was
cooled to -70•‹C and diniethylphenylsilyIlithium in THF (0.3 mL, 0.25 mmol) was added
dropwise. The reaction mixture was stirred on a vortex mixer in a custom built dewar for
10 min. MeLi (0.18 mL, 0.25 mmol) in Et20 was added via a syringe at
-78OC. The spectra were recorded immediately.
Preparation of 43 by Reaction of 17 and 28: CuCN (0.25 mL, 0.25
rnmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The solution was
cooled to -78OC and dimethylphenylsilyUithium in THF (0.6 mL, 0.5 mmol) added
dropwise. The reaction mixture was stirred on a vortex mixer in a custom built dewar for
20 min. In a separate vial, MeLi (0.36 mL, 0.5 mmol) was added to a THF solution of
CuCN (0.25 mL, 0.25 mmol) at -50•‹C. After 10 min of stirring, the solution of 17,
precooled to -78OC, was transferred via a cannula into cuprate 28, which was also
maintained at -78OC. The resulting deep red solution was stirred for 20 min before
recording the NMR spectrum
Preparation of (PhMe2Si)2(Me)CuLi2 (44) by Reaction of
PhMe2SiLi (25) and MeCu(CN)Li (16): CuCN (0.25 mL, 0.25 mmol) was
added to a 5 mm NMR tube, equipped with an argon inlet. The solution was cooled to - 50•‹C and MeLi (0.18 mL, 0.25 rnmol) in Et20 was added via a syringe. The reaction
mixture was stirred on a vortex mixer in a custom built dewar for 10 min.
DimethylphenylsilyIlithium (25) in THF (0.6 mL, 0.5 mrnol) was added dropwise to this
clear solution at -78OC. The resulting deep red solution was examined by NMR
immediately.
Preparation of 44 by Reaction of (PhMe2Si)2Cu(CN)Li2 (28) and
MeLi: MeLi in Et20 (0.36 mL, 0.5 mmol) was introduced dropwise via a syringe to a
solution of 28 (0.05 mmol, vide supra) at -78OC. The resulting deep red solution was
stirred for 20 min before recording the NMR spectrum.
Preparation of PhMezSi(Me)2CuLi2 (45) by Reaction of
PhMe2SiLi (25) MeLi and CuCN: CuCN (0.25 mL, 0.25 mmol) was added to a 5
rnm NMR tube, equipped with an argon inlet. The solution was cooled to -50•‹C. and
MeLi (0.36 mL, 0.5 mmol) in Et20 was added via a syringe. The reaction mixture was
stirred on a vortex mixer in a custom built dewar for 10 min. Upon addition of
dimethylphenylsilyIlithium (25) in THF (0.3 mL, 0.25 mmol) dropwise to this clear
solution at -78OC the reaction turned deep red. The NMR spectnun was recorded
immediately.
Preparation of 45 by Reaction of (PhMe2Si)3CuLi2 (29) and MeLi:
MeLi in Et20 (0.7 mL, 1.0 mmol) was introduced dropwise via a syringe to a solution
of 29 (1.0 mmol, vide supra) at -7g•‹C. The resulting deep red solution was stirred for 20
min before recording the NMR spectrum.
Preparation of 1 3 ~ NMR Samples for Reactions of 43 with 32:
PhMe2SiLi (0.3 mL, 0.25 mmol) was added dropwise at -85OC to a THF solution of
MeCu(CN)Li (0.25 mmol, vide supra) under argon. The resulting deep red solution was
stined for 10 min after which 32 (0.024 mL, 0.25 mmol) was added via a syringe. The
reaction was stirred for 10 min before recording the NMR spectrum.
Preparation of 1% NMR Samples for Reaction of 29 with 32: 32
(0.024 mL, 0.25 mmol) was added via a syringe at -85OC to a THF solution of
(PhMe2Si)3CuLi2,29 (0.25 mmol, vide supra). The reaction was stirred for 10 min
before recording the NMR spectrum.
Preparation of 3-trialkylsilyl cyclohexanone
Typical Procedure for Reactions of PhMe2SiLiICuCN Solutions
with 32, Synthesis of 33: PhMe2SiLi (1.25 mL, 1.0 mmol) was added dropwise at
-45OC to CuCN (0.089 g, 1.0 mmol) in THF (2 mL) under argon. The resulting deep red
solution was stirred for 0.5 h after which 32 (0.08 rnL, 0.82 mmol) was added via a
syringe. All reactions were stirred for a further 0.5 h and then quenched with satd.
NHqCV10% NH40H. The usual workup involved extraction of the organic phase with
Et20 (2 x 2 mL) and washing with brine (2 x 2 mL). The combined extracts were dried
over anhydrous MgS04 and concetrated in vacua Column chromatography (4: 1
hexanes:EtOAc) yielded 3-(dimethylphenylsily1)-cyclohexanone (33). The ratios of 1,2-
vs 1,4-addition products and the overall yields are reported in text. The 1~ NMR and IR
data for 33 matched those reported by Fleming et a ~ . 3 ~ c for this compound. 1 3 ~ ( 1 ~ )
(CDC13) 6 212.5 (C=O), 136.6 (ipso), 133.8, 129.2, 127.8, 42.3, 41.8, 29.7, 27.5,
26.0, -5.4 (SiCH3), -5.5 (SiCH3); MS m/e (rel. intensity) 232 (M+, 20), 217 (15), 189
(3 , 156 (22), 135 (100); Anal. calc. C14H200Si 232.1283 found 232.1282.
Typical Procedure for Reactions of PhMe2SiLiI CuBr*Me2S with
32: PhMe2SiLi (1.25 mL, 1.0 mmol) was added dropwise at -45OC to CuBpMe2S
(0.10 g, 0.5 mmol) in THF (2 mL) under argon. The resulting deep red solution was
stirred for 0.5 h after which 32 (0.04 mL, 0.41 mmol) was added via a syringe. All
reactions were stirred for a further 0.5 h and then quenched. Standard workup and the
spectral data for 33 are as reported above.
Typical Procedure for Reactions of PhMe2SiLi/MeLi/ CuCN
Solutions with 32, Synthesis of 33: PhMe2SiLi (1.25 rnL, 1.0 mmol) was added
dropwise at -70•‹C to a solution of MeCu(CN)Li [(1.0 mmol, prepared from the addition
of MeLi in EQO (0.7 mL, 1.0 mmol) and CuCN (0.089 g, 1.0 mmol in THF (2 mL) at
-50•‹C)] in THF (2 mL) under argon. The resulting deep red solution was stirred for 0.5 h
after which 32 (0.08 mL, 0.82 mmol) was added via a syringe. All reactions were
stirred for a further 0.5 h and then quenched. Standard workup and the spectral data for
33 are as reported above.
Preparation of 1-(Dirnethylphenylsily1)-cyclohex-2-en-1-01 (34):
Cyclohex-2-en-1-one (0.08 mL, 0.82 mmol) was added dropwise to a solution of
PhMe2SiLi (1.25 mL, 1.0 mmol) at'-45OC in THF (2 mL) under argon. The solution
turned yellow upon completion of addition. The reaction was stirred for 0.5 h and then
quenched with satd. NI-QCVlO% NI-QOH. The usual workup followed by column
chromatography (4: 1 hexanes:EtOAc) gave 62% of 34 as a colorless oil. IR (NaCl) 3460
(OH), 1440 (PhMe2Si) cm-l; 400 MHz IH NMR (CDC13) 6 7.2-7.7 (m, 5H, Ph), 5.9
(ddd, J = 10, 5, 3 Hz, lH, CH2HC=C), 5.7 (dt, J = 10, 3 HZ, lH, C=CH), 1.4-2.1
(m, 6H, ring CHz), 1.1 8 (bs, lH, OH), 0.38 (s, 3H, SiCH3), 0.36 (s, 3H, SiCH3);
1 3 ~ ( IH] (CDC13) 6 136.4 (ipso), 134.6, 130.4, 130.2, 129.2, 127.7, 64.2 (COH),
32.7,25.2, 17.5, -5.8 (SiCH3), -6.0 (SiCH3); MS d e (rel. intensity) 232 (M+, 5), 214
(M+-18,20), 199 (15), 135 (100); Anal. calc. C14H200Si 232.1284 found 232.1286.
Typical Procedure for Silylcupration with "PhMe2SiCuW
Reaction of PhMe2SiLi with 35a: 1-Octyne (0.24 g, 2.2 mmol) was added
dropwise to a solution of PhMe2SiLi (2.6 mL, 2.2 mmol) at -45OC in THF (3 mL) under
argon. The reaction was stirred for 0.5 h and then quenched with 2 ~ 2 0 (5 mL). It was
gradually warmed to room temperature. The usual workup yielded octpe- 12H (70%
incorporation as calculated from GC-MS analysis).
For 35a: MS m/e (rel. intensity) 95 (M+-15,30), 8 1 (100), 67 (52).
For 35b: MS m/e (rel. intensity) 96 (M+-15,25), 95 (17.5), 82 (loo), 81 (28),
68 (40).
Synthesis of 36 and 37: CuCN (0.197 g, 2.2 mmol) was placed in a flask
equipped with an argon inlet. The flask was repeatedly (3x) evacuated and purged with
argon. THF (5 mL) was injected, the reaction was cooled to -50•‹C and
dimethylphenylsilyllithium in THF (2.6 mL, 2.2 rnmol) added dropwise. The resulting
deep red solution was stirred for 0.5 h after which 1-Octyne (0.24 g, 2.2 mmol) was
added dropwise. The reaction was stirred for additional 0.5 h and then quenched with
H20 (5 rnL) for the preparation of 36s and 37a, whereas 2 ~ 2 0 (5 mL) was used as the
quenching agent for the generation of 36b. The reactions were warmed gradually to
room temperature. The usual workup followed by column chromatography (hexanes)
yielded the vinyl silanes which were analyzed by 1~ NMR and GC-MS analysis to
determine the amount of 2H incorporated. The ratios of 36 to 37 and the overall yields
are reported in the text.
For 36a: IR (NaCl) 1620,1440,1250, 1 120,995 cm- l ; 400 MHz IH NMR
(CDC13) 6 7.1-7.5 (m, 5H, Ph), 6.17 (dt, J = 18, 6 Hz, lH, HC=CSi), 5.85 (dt, J =
18, 1.5 Hz, lH, C=CHSi), 2.7 (tdd, J = 7, 6, 1.5 Hz, 2H, allylic), 1.2-1.42 (m, 8H,
CH2), 0.95 (t, J = 7 Hz, 3H, CH3), 0.3 (s, 6H, SiCH3); IH) (CDC13) 6 149.5
(C=CSi), 139.4 (C=CSi), 133.8 (ipso), 128.7, 127.6, 127.2, 36.8, 36.7, 31.7, 28.8,
28.6,22.5, 13.9 (SiCH3); MS m/e (rel. intensity) 246 (M+, 9), 23 1 (70), 161 (35). 135
(50), 121 (100); Anal. calc. Cl(jH26Si 246.1803 found 246.1809.
For 36b: 400 MHz IH NMR (CDC13) 6 7.1-7.5 (m, 5H, Ph), 5.85 (brt J = 1.5
Hz, lH, C=CHSi), 2.7 (td, J = 7, 1.5 Hz, 2H, allylic), 1.2-1.42 (m, 8H, CH2), 0.95
(t, J = 7 Hz, 3H, CH3), 0.3 (s, 6H, SiCH3); MS m/e (rel. intensity) 247 (M+, 12), 232
(loo), 162 (go), 148 (40), 135 (60), 122 (go), 121 (80); Anal. calc. C16H25DSi
247.1866 found 247.1874.
For 37a: 400 MHz IH NMR (CDC13) 6 5.9 (dt, J = 7, 1.5 Hz, 2H, vinyl); MS
m/e (rel. intensity) 246 (M+, 43,231 (43, 161 (33, 135 (loo), 121 (35).
Gilman Tests. All cuprates prepared in 10 mm NMR tubes were subjected to
Gilman test. An equal volume of Michler's ketone (1% solution) in dry benzene was
added to the cuprate at O•‹C. H20 (1 mL) was introduced after 10 min and the reaction
was allowed to warm to room temperature. After vigorous stirring, a 2% solution of I2 in
glacial acetic acid was added dropwise. A persistent blue color in the organic layer is
considered a positive test (confirmation of free
PREPARATION OF TRIALKYLSTANNYLCUPRATES
-1%n NMR sample preparation
Preparation of MefSnLi (47): According to the procedure of Still et aL49a
MeLi (7.15 mL, 10.0 mmol) was added dropwise at -45OC to MegSn2 (3.27 g, 10.0
mmol) in THF (10 mL) under argon. The resulting pale yellow solution was stirred for
0.5 h and titrated according to procedure of ~ilrnanggb before being used in the
preparations of cuprates.
Generation of stannylcuprates from CuCN
Preparation of MesSnCu(CN)Li (48): CuCN (0.09 g, 1.0 mmol) was
placed in a 10 mm NMR tube, equipped with an argon inlet. The tube was repeatedly
(3x) evacuated and purged with argon. The reaction was cooled to -70•‹C and
trimethyltinlithium in THF (2.0 mL, 1.0 rnmol) was added dropwise. The orange slurry
was stirred on a vortex mixer at -78OC in a custom built dewar for 20 min. The NMR
spectrum was then immediately recorded.
Preparation of (Me3Sn)2Cu(CN)Li2 (51): A THF solution of Me3SnLi
(2.0 mL, 1.0 mmol) was added to a 10 mm NMR tube containing a THF solution of
Me3SnCu(CN)Li at -78OC. The orange colored solution was stirred for 20 min at -78•‹C
prior to examination by NMR and IR spectroscopy.
Preparation of (Me3Sn)3CuLi2 (50): To a THF solution of
(Me3Sn)2Cu(CN)Li2 (1.0 mmol) prepared as outlined above was added a THF solution
of Me3SnLi (2.0 mL, 1.0 mmol) at -78OC. The reaction was stirred for 20 min at -78OC
before examination by NMR and IR spectroscopy.
Preparation of (Me3Sn)3CuLi2 (50) with excess 47: To a THF
solution of (Me3Sn13CuLi2 (1.0 mmol, vide supra) was added a THF solution of
Me3SnLi (2.0 mL, 1.0 mmol) at -78OC. The reaction was stirred for 20 min at -78OC
before examination by NMR and IR spectroscopy.
Regeneration of (Me3Sn)2Cu(CN)Li2 (51) from 50: CuCN (0.045 g,
0.05 mmol) was added to the stannylcuprate 50. The reaction was stirred for 20 min at
-78OC before examination by NMR spectroscopy.
Regeneration of MesSnCu(CN)Li (48) from 51: CuCN (0.09 g, 1.0
mmol) was added to the NMR tube containing stannylcyanocuprate 51. The reaction was
stirred for 20 min at -78OC and then examined by NMR spectroscopy.
Generation of Stannylcuprates from CuBroMe2S
Preparation of Me3SnCu (49): CuBroMe2S (0.205 g, 1.0 mmol) was
placed in a 10 mm NMR tube, equipped with an argon inlet. The tube was repeatedly
(3x) evacuated and purged with argon. The solution was cooled to -78OC and
trimethyltinlithium in THF (2.0 mL, 1.0 mmol) was added dropwise. A red slurry was
obtained. The reaction mixture was stirred on a vortex mixer at -78OC in a custom built
dewar for 20 min. The NMR spectrum was immediately recorded without filtration.
Preparation of'(Me3Sn)2CuLi (52): A THF solution of Me3SnLi (2.0 mL,
1.0 mmol) was added to a 10 mm NMR tube containing a THF solution of Me3SnCu 49
(1.0 mmol, vide supra) at -78OC. The resultant deep red colored solution was stirred for
20 min at -78OC prior to examination by NMR spectroscopy.
Preparation of (Me3Sn)3CuLi2 (50): To a THF solution of
(Me3Sn)2CuLi (1.0 mmol) prepared as outlined above, was added a THF solution of
Me3SnLi (2.0 mL, 1.0 mmol) at -78OC. The reaction mixture was stirred for 20 min at -
78OC before examination by NMR spectroscopy.
Preparation of (Me3Sn)3CuLi2 (50) with excess Me3SnLi (47): To a
THF solution of (Me3Sn)3CuLi2 (50, 1.0 mmol) prepared as outlined above, was added
a THF solution of Me3SnLJ (2.0 mL, 1.0 rnmol) at -78OC. The reaction mixture was
stirred for 20 min at -78OC before examination by NMR spectroscopy.
Regeneration of (Me3Sn)zCuLi (52) from 50: CuBroMe2S (0.10 g, 0.5
mmol) was added to 1.0 mmol of the solution generated by mixing Me3SnLi (47) and
CuBr.Me2S in a 3:l ratio. The reaction mixture was stirred for 20 min at -78OC before
examination by NMR spectroscopy.
Regeneration of Me3SnCu (49) from 52: CuBroMe2S (0.205 g, 1.0
mmol) was added to the NMR tube containing stannylcuprate 52 (1.0 mmol). The
reaction was stirred for 20 min at -78OC and then examined by NMR spectroscopy.
Generation of Mixed Stannylcuprates
Preparation of Me3Sn(Me)Cu(CN)Li2 (56) by Reaction of Me3SnLi (47)
and MeCu(CN)Li (16): CuCN in THF (0.09 g, 1.0 mmol) was added to a 10 mm
NMR tube, equipped with an argon inlet. The solution was cooled to -50•‹C and MeLi
(0.71 mL, 1.0 mmol) in Et20 was added via a syringe. The reaction mixture was stirred
on a vortex mixer in a custom built dewar for 10 min. Trimethyltinlithium in THF (2.0
mL, 1.0 mmol) was added dropwise to this clear solution at -78OC. The reaction turned
clear yellow in color. The spectra were recorded immediately.
Preparation of 56 by Reaction of Me3SnCu(CN)Li (48) and MeLi:
MeLi in Et20 (0.7 1 mL, 1.0 mmol) was added to the orange solution of 48 (1.0 mmol,
prepared as above) at -78OC. The clear yellow solution was stirred on a vortex mixer at - 78OC in a custom built dewar for 20 min. The NMR spectrum was then immediately
recorded.
Preparation of 56 by Reaction of (Me3Sn)2Cu(CN)Li2 (51) and
Me2Cu(CN)Li2 (17): CuCN (0.09 g, 1.0 mmol) was added to a flask, equipped with
an argon inlet. The solution was cooled to -50•‹C where MeLi (1.4 mL, 2.0 mmol) was
introduced dropwise generating a clear solution in 10 min. It was then transferred via a
cannula into cuprate 51 (1.0 mmol, prepared as above) which was also maintained at
below -78OC. The resulting yellow solution was stirred for 20 min before recording the
NMRspec-
1~ and 1 3 ~ NMR Sample Preparation
Generation of Stannylcuprates from CuCN
Preparation of 48: A THF solution of Me3SnLi (0.5 rnL, 0.25 mmol) was
added to a 5 mm NMR tube, equipped with an argon inlet. The solution was cooled to
-78OC and a solution of CuCN in THF (0.25 mL, 0.25 mmol) added dropwise. The
reaction mixture was stirred on a vortex mixer in a custom built dewar for 20 rnin. The
spectra were recorded immediately. Inverse addition of the reagents gave similar spectral
results.
Preparation of 51: A THF solution of Me3SnLi (0.5 mL, 0.25 -01) was
added to a 5 mm NMR tube, equipped with an argon inlet. The solution was cooled to
-78OC and a solution of CuCN in k (0.125 mL, 0.125 -01) added dropwise. The
reaction mixture was stirred on a vortex mixer in a custom built dewar for 20 min. The
spectra were recorded immediately. Similar spectral results were obtained when the order
of mixing of reagents was reversed.
Preparation of 50: A THF solution of Me3SnLi (0.5 mL, 0.25 mmol) was
added to a 5 mm NMR tube, equipped with an argon inlet. The solution was cooled to
-78OC and a solution of CuCN in THF (0.08 mL, 0.08 mmol) added dropwise. The
reaction was stirred on a vortex mixer in a custom built dewar for 20 min. The spectra
were recorded immediately. Inverse addition of the reagents gave similar spectral results.
Preparation of 50 with excess 47: A THF solution of Me3SnLi (0.5 mL,
0.25 -01) was added to a solution of stannylcuprate 50 at -78OC. The reaction was
stirred on a vortex mixer in a custom built dewar for 20 min. The spectra were recorded
immediately. Inverse addition of the reagents gave similar spectral results.
Preparation of 48 for IH NMR analysis: The above THF solution of
CuCN (0.25 mL, 0.25 mmol) was added to a 5 mm NMR tube, equipped with an argon
inlet. The solution was cooled to -7g•‹C and trimethyltinlithium in THF-dg (0.5 mL, 0.25
mmol) added dropwise. The reaction mixture was stirred on a vortex mixer in a custom
built dewar for 20 min. The spectra were recorded immediately.
Preparation of 50 and 51 for 1~ NMR analysis: These solutions were
prepared in THF-dg as described above for 48.
Generation of Stannylcuprates from CuBr.Me2S
Preparation of Me3SnCu (49): CuBr.Me2S (0.25 mL 0.25 mmol) was
added to a 5 mm NMR tube, equipped with an argon inlet. The reaction was cooled to
- 7g0C and trimethyltinlithium in THF (0.5 mL, 0.25 mmol) added dropwise. The
solution was stirred on a vortex mixer in a custom built dewar for 20 min. A deep red
sluny was obtained. The spectra were recorded without filteration.
Preparation of (Me3Sn)~CuzLi (53): CuBr.Me2S (0.17 mL 0.166 mmol)
was added to a 5 mm NMR tube, equipped with sn A+ZF i n k t rp.action was cooled
to -78OC and trimethyltinlithium in THF (0.5 mL, 0.25 rnmol) was added dropwise. The
solution was stirred on a vortex mixer in a custom built dewar for 20 min. The slurry
was used as such.
Preparation of (Me3Sn)3CuLi2 (50): Trimethyltinlithium in THF (0.5 mL,
0.25 mmol) was added dropwise to a solution of 53 (0.25 mmol) at -78OC. The solution
was stirred on a vortex -mixer in a custom built dewar for 20 rnin. The spectra were
recorded immediately.
Generation of Mixed Stannylcuprates
Preparation of 56 by Reaction of 47 and 16: CuCN in THF (0.4 mL,
0.4 mrnol) was added to a 5 mm NMR tube, equipped with an argon inlet. The solution
was cooled to -50•‹C and MeLi (0.3 mL, 0.4 mmol) in Et20 was added via a syringe.
The reaction mixture was stirred on a vortex mixer in a custom built dewar for 10 rnin.
Trimethyltinlithium in THF (1.0 mL, 0.4 mmol) was added dropwise to this clear
solution at -78OC. The reaction turned yellow in color. The spectra were recorded
immediately.
Preparation of 56 by Reaction of 17 and 51: CuCN in THF (0.4 mL,
0.4 mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The solution
was cooled to -78OC and trimethyltinlithiurn in THF (1.0 mL, 0.4 mmol) added
dropwise. The reaction mixture was stirred on a vortex mixer in a custom built dewar for
20 min. In a separate vial, 17 (prepsred as above) precooled to -78OC, was transferred
via a cannula into cuprate 51, which was &so mniihraineci a1 -Z•‹C. The resulting yellow
solution was stirred for 20 min before recording the NMR spectrum.
Preparation of Me3Sn(Me)zCuLi2 (57) by Reaction of 48 and
MeLi: CuCN in THF (0.4 mL, 0.4 mmol) was added to a 5 mm NMR tube, equipped
with an argon inlet. The solution was cooled to -78OC and Me3SnLi (1.0 mL, 0.4 rnmol)
in THF was added via a syringe. The reaction mixture was stirred on a vortex mixer in a
custom built dewar for 10 min. MeLi (0.6 mL, 0.8 mmol) in EQO was added dropwise
to this orange suspension at -78OC. The reaction turned clear yellow in color. The spectra
were recorded immediately.
Preparation of 57 by Reaction of 56 and MeLi: CuCN in THF (0.4 mL,
0.4 mmol) was added to a 5 mm NMR tube, equipped with an argon inlet. The solution
was cooled to -50•‹C and MeLi (0.3 mL, 0.4 mmol) in Et20 was added via a syringe.
The reaction mixture was stirred on a vortex mixer in a custom built dewar for 10 min.
Trimethyltinlithium in THF (1.0 mL, 0.4 mmol) was added dropwise to this clear
solution at -78OC. The reaction turned yellow in color. MeLi (0.3 mL, 0.4 mmol) in Et20
was then added via a syringe to yield a clear yellow solution. The spectra were recorded
immediately.
Reaction of Slwith MeLi: MeLi in Et20 (0.15 mL, 0.2 mmol) was added
dropwise to a THF solution of 51 (0.4 mmol) at -78OC. The solution was stirred on a
vortex mixer in a custom built dewar for 20 min. The spectra were recorded immediately.
Reaction of 50 with 56: A THF solution of (Me3Sn)3CuLi2 (0.25 mmol)
was prepared in a 5 mm NMR tube as described above. The solution was cooled to
-78OC and a solution of Me3Sn(Me)Cu(CN)Li2 in THF (0.25 mmol) added dropwise.
The reaction mixture was stirred on a vortex mixer in a custom built dewar for 20 min.
Reaction of 58 with 17, generation of 60: Bu3SnH (0.13 mL 0.5 mmol)
was added dropwise at -85OC to Me2Cu(CN)Li2 (0.5 mmol, vide supra). NMR was
recorded on the resulting yellow solution after stirring for 10 rnin.
Reaction of 58 with 17, generation of 61: Bu3SnH (0.13 mL 0.5 mmol)
was added dropwise at -85OC to the above solution. After the evolution of H2 subsided
(5 min) NMR was recorded on the resulting yellow solution.
Preparation of 34rialkylstannyl cyclohexanone
Typical Procedure for Reactions of Me3SnLi/MeLi/CuCN Solutions
with 32, Synthesis of 70: Me3SnLi (2.0 mL, 1.0 mmol) was added dropwise at
-78OC to a solution of MeCu(CN)Li (16,l.O mrnol, generated fiom.the reaction of MeLi
in Et20 (0.75 mL, 1.0 mmol) and CuCN (0.09 g, 1 mmol) in THF (2 mL) under argon.
The resulting solution was stirred for 0.5 h after which 32 (0.08 mL, 0.82 mmol) was
added via a syringe. The reaction was stirred for a further 0.5 h and then quenched with
satd. NH4Cl/10% NH40H. Workup involved extraction of the organic phase with Et20
(2 x 2 mL) and washing with brine (2 x 2 mL). The combined extracts were dried over
anhydrous MgS04 and concetrated in vacuo. Column chromatography (4: 1
hexanes:EtOAc) yielded 3-(trimethylstanny1)-cyclohexanone (70) in > 90% isolated
yield. The IH NMR and IR data for 70 matched those reported by for this
compound. 1 3 ~ ( 1 ~ ) (CDC13) 6 212.2 (C=O), 45.8,42.1, 30.8, 29.4, 25.2, -1 1.7
(SnCH3); MS m/e (rel. intensity) 246 (M+-15,20); Anal. calc. CgHlgOSn 246.1283
found 246.1282.
Reaction of Bu3SnH/ MezCu(CN)Liz with 32, Synthesis of 71:
Bu3SnH (0.27 mL, 1.0 mmol) was added dropwise at -78OC to a solution of
Me2Cu(CN)Li2 (17.1.0 -mo ly generated from the reaction of MeLi in Et20 (0.75 mL,
1.0 mmol) and CuCN (0.045 g, 0.5 mmol) in THF (2 mL) under argon. The resulting
yellow solution was stirred for 0.5 h after which 32 (0.082 mL, 0.82 rnmol) was added
via a syringe. The reaction was stirred for a further 0.5 h and then quenched with satd.
NH4Cl/10% NUOH. The usual workup involved extraction of the organic phase with
EQO (2 x 2 mL) and washing with brine (2 x 2 mL). The combined extracts were dried
over anhydrous MgS04 and concetrated in vacuo. Column chromatography (4:l
hexanes:EtOAc) yielded 3-(tributylstanny1)-cyclohexanone (71) in > 96% isolated yield.
The IH NMR and IR data for 71 matched those reported by ~ t i 1 1 ~ 8 ~ for this compound.
1 3 ~ ( 1 ~ ] (CDC13) 6 211.8 (C=O), 46.3, 42.1, 31.0, 30.0, 29.1, 27.3 (Cg), 13.4, 8.0;
MS m/e (rel. intensity) 331 (M+-56,70).
Preparation of BujSnH (58): Reduction of Bu3SnCI with L ~ A I H ~ ~ ~ gave
Bu3SnH in 82% chemical yield and 97% purity as measured by gas chromatographic
analysis after distillation (bp 4g•‹C @ 0.05 mm Hg). A modified work-up pmcedure was
employed that involved the transfer of the supernatant from the aqueous layer via a
canula. Removal of EQO followed by addition of n-hexanes in vacuo concentration and
distillation under reduced pressure gave oxide-free Bu3SnH that could be stored in the
freezer without noticable decomposition for several months.
Preparation of Trialkylstannyllithium and Quenching with Methyl Iodide,
Table IV:
Entry 1: Bu3SnLi was prepared according to the procedure of ~ozaki?O
Stannous chloride ( 0.6 g, 3.16 mmol) was stirred in 5 mL of dry THF at O•‹C and then
n-BuLi (3.65 mL, 9.48 mmol) was added dropwise over 30 rnin. To an aliquot of the
deep red solution was added excess Me1 in THF at O•‹C. The reaction was stirred for 15
min then quenched by addition of NH4Cl solution and the separated organic layer
analyzed by gas chromatography with dodecane as an internal standard. Bu3SnMe was
formed in 33% yield.
Entry 2: A solution of Bu3SnC1( 4.0 g, 12.5 mmol) in 25 mL of THF was
added to a suspension of Li clippings (0.35 g, 50 mmol) in 20 mL of THF at O•‹C
according m the procedure of ~oloski.~gb The reaction was stirred overnight at this
temperature and the green solution treated with excess MeI. The yield of Bu$bMe was
48% as determined by GC against an internal standard (dodecane).
Entry 3: A Li dispersion (1.4 g, 200 mmol) in n-hexane was prepared in a
preweighed Schlenk tube under argon. THF (75 mL) was added, followed by Me3SnC1
(10 g, 50 mrnol) in 50 mL of THF. The reaction was stirred at O•‹C for 8 h. An aliquot
was quenched with excess MeI. The yield of Me3SnLi was calculated from the amount
of MeqSn detected by gas chromatographic analysis vs decane as an internal standard.
Entry 4: n-BuLi (0.38 mL, 1.0 mrnol), was added to a stirred solution of
Bu3SnH (0.291 g, 1.0 mmol) in 5 mL of THF at O•‹C. The reaction was quenched with
Me1 after 15 min. Gas chromatographic analysis revealed BaSn as the major product
accompanied by only 4% of Bu3SnMe (dodecane as internal standard).
Entry 5: Prep&tion of BugSnLi by the mmed procedure of ~till49a
involved addition of BuLi (1.0 mL, 2.5 mrnol) to an efficiently stirred solution of
diisopropyl axnine (0.35 mL, 2.5 mmol) in 5 mL of THF at -lO•‹C. The reaction was
stirred for 30 min then 3SnH (0.72 g, 2.5 mrnol) was added at -30•‹C. The reaction
turned pale green at this point. The yield of Bu3SnMe was 90% as determined by gas
chromatographic analysis with dodecane as an internal standard.
Entry 6: To an efficiently stirred solution of Me3SnSnMe3 (1.63 g, 5.0 mmol)
in 20 mL of THF was added MeLi (3.6 mL, 5 mmol) while maintaining the temperature
below -40•‹C. After 20 min the reaction was quenched with excess Me1 and the yield
(go%) of MeqSn was calculated as before.
Reaction of 1-Decyne (91) with Bu3SnAIEt2 (90); Representative
Procedures for the Preparation of 2-Tributylstannyl-1-alkenes (92a, 92b,
93a and 93b): The following three procedures are representative of the addition of 91
to 90 (Table V).
(a) Using SnC12 and n-BuLi (Method A): n-BuLi (1 1.4 mL, 28.44
mmol) was added dropwise to a solution of stannous chloride (1.8 g, 9.48 mmol) in 15
mL of dry THF at O•‹C. After stirring for 30 min, diethylaluminum chloride (9.48 mL,
9.48 mmol) was added and the reaction stirred for a further 0.5 h. 1-Decyne (0.437 g,
3.16 mmol) and catalyst (0.3 mmol) were then added at -30•‹C. The reaction was stirred
for 3 h after which it was allowed to warm to O•‹C. Normal workup gave vinyl stannanes
92a and 93a in the ratios shown in the Table 2.
(b) Using B U ~ S ~ H and LDA (Method B): Bu3SnH (1.45 g, 5 mmol)
was added to 5 mL of THF containing lithium diisopropyl amide (prepared by dropwise
addition of n-BuLi (2 mL, 5 mmol) to an efficiently stirred solution of diisopropyl amhe
(0.70 mL, 5 mmol) in 5 mL of THF at -1(TC) while maintaining the temperature below - 30•‹C. After stirring for 0.5 h, Et2AICI (5 mL, 5 mmol) was added dropwise at -30•‹C.
The clear solution was further stirred for 0.5 h then 1-decyne (0.22 g, 1.6 mmol) in 5
mL of THF was added dropwise followed by the addition of a catalyst (0.16 rnmd). The
reaction turned orange at this time. After 3 h the reaction was warmed to O•‹C and
subjected to normal workup.
(c) General Procedure for slow Addition Reactions (Method C): To
3.2 mmol of lithium diisopropyl amide in 5 mL of THF was added Bu3SnH (0.87 g, 3
mmol) in 12 mL THF at -30•‹C. After stirring for 0.5 h, Et2AlCI (2.8 mL, 2.8 mmol) in
hexane was added dropwise. After stining for 0.5 h, one half of the 1-decyne to be
reacted (0.32 g, 2.3 mmol) was added in 15 mL of THF. This was followed by addition
of CuCN (0.022 g, 0.23 mmol) in 5 mL of THF. The remainder of the alkyne was
added over 0.75 - 1.0 h. The reaction was stirred at -30•‹C for 3 h then warmed to O•‹C
and subjected to normal workup.
The experimental results of the addition of 91 to 90 are summarized in Table V,
and the products gave the following spectral data
2-Tributylstannyl-1-decene (92a, Scheme 39): 1~ NMR (CDC13) 6
0.82 (t, 3H, CH3, J = 7.5 Hz), 6 0.88 (t, 9H, CH3, J = 7.0 Hz), 1.2-1.4 (my 22H,
CHz), 1.45- 1.6 (my 8 ~ ; CH2), 2.23 (ddt, 2H, C=CCH2, J = 7.56 Hz, 1.45 Hz, 1 .O4
Hz), 5.09 (dt, IH, C=CHy J = 2.9 HZ, 1.04 HZ, 3 ~ ~ n - ~ = 64 Hz), 5.66 (dt, IH,
C=CHy J = 2.9 Hz, 1.45 Hz, 3 ~ ~ n - ~ = 140 Hz). Allylic region decoupled: 6 2.21 (s),
5.09 (d, 1Hy C=CHy J = 2.9 Hz), 5.66 (d, lHy GCH, J = 2.9 HZ); 1 3 ~ NMR 156.0
(C=CSnBu3), 124.5 (C=CH2), 41.4, 32.0, 29.7, 29.5, 29.3, 29.1, 27.4, 22.6, 13.6,
9.7; 19sn NMR = - 45.6; GUMS, d e (relative intensity), 373 (M+ - 56,37.5),29 1
(100.0). Anal Calcd for C18H37Sn: 373.19 17. Found: 373.1920; for C 14H2gSn:
317.1291. Found: 317.1288 .
E-Tributylstannyl-1-decene (93a): IH NMR (CDC13) 6 0.8 (t, 3H, CH3,
J = 5.7 Hz), 0.88 (t, 9H, CH3, J = 7.0 Hz), 1.2-1.4 (my 22H, CH2), 1.45-1.6 (my 8H,
CH2), 2.2 (dq, 2H, C=CCH2, J = 5.5 Hz, 2.0 Hz), 5.85 (dt, lH, C=CH, J = 18.0 Hz,
2.0 Hz, 2 ~ ~ n - ~ = 60 Hz), 5.96 (dt, IH, C=CH, J = 18.0 HZ, 5.5 HZ, 3 ~ ~ ~ - ~ = 47
Hz). Allylic region decoupled, 2.2 (s, 2H), 5.85 (d, lHy C=CH, J = 18.0 Hz), 5.96 (d,
lH, C=CH, J = 18.0 Hz); 13c NMR 155.3 (C=CSnBug), 121.4 (C=CH), 42.2, 31.9,
29.5, 29.3, 28.2, 27.3, 24.4, 22.7, 13.6, 10.0; 1 1 9 ~ n NMR = - 50.1; GC/MS, m/e
(relative intensity), 373 (M+ - 56,31.25), 291 (100.0). Anal Calcd for ClgH37Sn:
373.1917. Found: 373.1916; for C14H2gSn: 317.1291. Found: 317.1290.
Independent preparation of 93a (Scheme 39): To 1-Decyne (0.138 g,
1.0 mmol) in 5 mL of n-hexanes DIBALH (1 mL, 1.0 mmol) was added dropwise with
stirring . The reaction was refluxed for 2 h after which time the solvent was removed
under vacuum. THF (5 mL) was then added and the solution was cooled to -78OC
whereupon n-BuLi (0.8 mL, 2.0 mmol) and HMPA (2 rnL) were added. The reaction
was stirred for 0.5 h then Bu3SnC1(0.65 g, 2.0 mrnol) was added and the reaction
further stirred for an hour at -78OC and 2 h at 0-C. Normal workup gave 0.32 g (76%) of
(93a). BuqSn was the other product detected.
Preparation of 92b and 93b (Scheme 39): The reaction was conducted as
described above (Method B), but was quenched by stirring with Z H C ~ for 0.5 h and then
processed in the usual fashion.
2-1-Deuterio-2-tributylstannyl-1-decene (92b): 1~ NMR (CDC13) 6
0.88 (t, 9H, CH3, J = 7.0 Hz), 0.9 (t, 3H, CH3, J = 7.7 Hz), 1.2-1.4 (my 22H, CH2),
1.45-1.6 (my 8H, CH2), 2.24 (dt, 2H, C=CCH2, J = 7.66 Hz, 1.5 Hz), 5.66 (t, lH,
C=CH, J = 1.5 Hz, 3 ~ ~ n - ~ = 140 Hz); GC/MS, m/e (relative intensity), 374 (M+ - 56,
97). Anal Calcd for C I ~ H ~ ~ S ~ ~ H : 318.1354. Found: 318.1350.
2-2-Deuterio-1-tributylstannyl-1-decene (93b): 1~ NMR (CDC13) 6
0.88 (t, 9H, CH3,J= 7.0 HZ), 0.9 (t, 3H, CH3, J = 5.7 Hz), 1.2-1.4 (my 22H, CH2), t
1.45- 1.6 (m, 8H, CH2), 2.2 (dt, 2H, C=CCH2, J = 5.5 Hz, 3.0 Hz), 5.85 (t, IH,
C=CHy J = 3.0 Hz, 2 ~ ~ n - ~ = 60 Hz); GC/MS. m/e (relative intensity), 374 (M+ - 56,
100); Anal Calcd for c14F12gsn2~: 318.1354. Found: 318.1361.
Reaction of 1-Nonyne (95) with [(Bu3Sn-9-BBN0Me)-]Li+ (94);
Representative Procedures for the Preparation of 2-Tributylstannyl-l-
alkenes (96a, 97a, 96b and 97b): The following procedure is representative of the
addition of 95 to 94, (Scheme 40).
Bu3SnH (1.5 g, 5.8 rnrnol) in THF (10 mL) was added to 5 mL of THF
containing lithium diisopropyl amide (prepared by dropwise addition of n-BuLi (2.5 mL,
6.0 mmol) to an efficiently stirred solution of diisopropyl amine (0.84 mL, 6.0 mmol) in
5 mL of THF at -lO•‹C) while maintaining the temperature below -30•‹C. After stining for
0.5 h, B-methoxy-9-borabicyclo[3.3.l]nonane (1 1.6 mL, 5.8 mmol) was added
dropwise at -78OC. The clear solution was warmed to 0•‹C and further stirred for 0.5 h .
The reaction was cooled to -78OC and then 1-nonyne (0.38 g, 3.0 mmol) in 5 mL of
THF was added dropwise followed by the addition of a catalyst (0.3 mmol). The
reaction turned orange at this point. After 3 h at -65OC, a coupling agent was added.
Stirring was continued for 1 h and then the reaction was quenched and oxidized with
NaOH (15 mL and H202 (5 mL). The resulting solution was filtered through celite and
then subjected to column purification. Spectral data for the various products are listed
below:
2-Tributylstannyl-1-nonene (96a, Scheme 40): 1~ NMR (CDC13) 6
0.82 (t, 3H, CH3, J = 7.5 Hz), 6 0.88 (t, 9H, CH3, J = 7.0 Hz), 1.2-1.4 (m, 20H,
CH2), 1.45- 1.6 (m, 8H, CH2), 2.23 (ddt, 2H, C=CCH2, J = 7.56 Hz, 1.45 Hz, 1 .O4
HZ), 5.09 (dt, IH, C=CH, J = 2.9 HZ, 1.04 HZ, 3 ~ ~ n - H = 64 HZ), 5.66 (dt, IH,
C=CH, J = 2.9 Hz, 1.45 Hz, 3 ~ ~ n - H = 140 Hz); 13c NMR 156.0 (C=CSnBu3),
124.5 (C=CH2), 41.4, 32.0, 29.5, 29.3, 29.1, 27.4, 22.6, 13.6, 9.7; 119sn NMR = - 45.6; GC/MS, m/e (relative intensity), 361 (M+ - 56,373,291 (100.0).
E-Tributylstannyl-1-nonene (97a): IH NMR (CDC13) 6 0.8 (t, 3H, CH3,
J = 5.7 Hz), 0.88 (t, 9H, CH3, J = 7.0 Hz), 1.2-1.4 (m, 20H, CH2), 1.45-1.6 (m, 8H,
CH2), 2.2 (dq, 2H, C=CCH2, J = 5.5 Hz, 2.0 Hz), 5.85 (dt, lH, C=CH, J = 18.0 Hz,
2.0 Hz, 2 ~ ~ n - ~ = 60 HZ), 5.96 (dt, IH, C=CH, J = 18.0 HZ, 5.5 HZ, 3 ~ ~ ~ - ~ = 47
Hz); 13c NMR 155.3 (C=CSnBu3), 121.4 (C=CH), 42.2, 31.9, 29.3, 28.2, 27.3,
24.4,22.7, 13.6, 10.0: 119sn NMR = - 50.1; GC/MS, m/e (relative intensity), 361
(M+ - 56, 31-25), 291 (100.0).
2-1-Deuterio-2-tributylstannyl-1-nonene (96b): 1~ NMR (CDC13) 6
0.88 (t, 9H, CH3, J = 7.0 Hz), 0.9 (t, 3H, CH3, J = 7.7 Hz), 1.2-1.4 (m, 20H, CH2),
1.45- 1.6 (m, 8H, CH2), 2.24 (dt, 2H, C=CCH2, J = 7.66 Hz, 1.5 Hz), 5.66 (t, lH,
C=CH, J = 1.5 Hs 3 ~ ~ n - ~ = 140 Hz); GC/MS, m/e (relative intensity), 362 (M+ - 56,
97).
2-2-Deuterio-1-tributylstannyl-1-nonene (97b): 1~ NMR (CDC13) 6
0.88 (t, 9H, CH3, J = 7.0 Hz), 0.9 (t, 3H, CH3, J = 5.7 Hz), 1.2-1.4 (m, 20H, CH2),
1.45-1.6 (m, 8H, CH2), 2.2 (dt, 2H, C=CCH2, J = 5.5 Hz, 3.0 Hz), 5.85 (t, lH,
C=CH, J = 3.0 Hz, 2 ~ ~ n - ~ = 60 Hz); GC/MS, m/e (relative intensity), 362 (M+ - 56,
100).
BuqSn and Bu3SnSnBu3 were identified by NMR and gas chromatogrC$k
comparison with authentic samples.
In the absence of a catalyst but o t h d s e under the same conditions the reaction
did not produce vinylstannanes.
General Procedure for the Preparation of Disubstituted
Vinylstannanes (Scheme 39): 1-Decyne (0.69 g, 5.0 mmol) in 5 mL of THF was
added to a solution of 90 at -30•‹C (10 rnmol, vide supra) in 20 mL of THF. This was
followed by addition of CuCN (0.045 g, 0.5 rnmol) at which point the reaction turned
orange. After stirring for 3 h at -30•‹C it was warmed to O•‹C (wine red in color) and
subjected to two separate sets of conditions as described below:
(a) Transmetallation with n-BuLi: Preparation of 92c, 92d and 93d: . n-BuLi (2.4 mL, 6.0 mmol) was added at -78OC to 1,2-dimetallo- 1-akene (prepared as
described above). After stirring for 30 min an excess of ally1 bromide (92c, 2x) or Me1
(2x) in HMPA (5 mL, 92d and 93d) was added. The reaction was left to stir overnight.
Usual workup followed by silica gel chromatography (n-hexanes as eluant) gave the
desired products in >98% stereoisomeric purity.
5-Tributylstannyl-4(Z)-1-tridecadiene (92c): 1.6 g (68%); 1~ NMR
(CDC13) 6 0.88 (t, 12H, CH3, J = 7.0 Hz), 1.2- 1.4 (m, 22H, CH2), 1 A < - 1 . 6 {rr, W,
CH2), 2.23 (dt, 2H, C=CCH2, J = 7.0 HZ, 1.5 Hz), 2.73 (ddt, 2H, C=CCH2C=C, J =
7.0 Hz, 2.0 Hz, 1.5 Hz), 4.99 (ddt, lH, HC=CHcis, J = 10.0 Hz, 2.0 Hz, 1.5 Hz),
5.02 (dq, 1 H, HC=CHtrans, J = 17.0 Hz, 2.0 Hz), 5.8 (ddt, 1 H, HC=CH2, J = 17.0
Hz, 10.0 HZ, 6.0 HZ), 5.96 (ddt, lH, C=CHCH2, J = 7.0 Hz, 2.0 Hz, 1.5 HZ, 3~Sn-H
= 138 Hz); l 3 c NMR 6 156.0 (C=CSnBug), 148.0 (C=CH), 139.5 (HC=CH2), 138.0
(C=CH2), 115.0 (C=CCH2C=C), 41.4, 32.0, 29.6, 29.3, 29.2, 29.1, 27.4, 22.6,
13.6,9.7; 19sn NMR = - 52.1; GC/MS, d e (relative intensity), 413 (M+ - 56, 100);
Anal Calcd for C21Hq3Sn: 413.1354. Found: 413.1361.
2-Tributylstannyl-l(Z)-undecene (92d): 0.85 g (38%); IH NMR
(CDC13) 6 0.88 (t, 12H, CH3, J = 7.0 Hz), 1.2- 1.4 (m, 22H, CH2), 1.45-1.6 (m, 8H,
CH2), 1.96 (d, 3H, CH3, J = 7.0 Hz), 2.23 (dt, 2H, C=CCH2, J = 7.0 Hz, 1.9 Hz),
6.1 (qt, IH, C=CH, J = 7.0, 1.9 HZ, 3~Sn-H = 138 HZ); l 3 ~ NMR 6 145.2
(C=CSnBu3), 134.0 (C=CH), 40.8, 30.7, 29.3, 29.2, 27.8, 26.8, 24.2, 15.5, 14.0,
10.2; 19sn NMR = - 45.0; GC/MS, d e (relative intensity), 387 (M+ - 56,100); Anal
Calcd for ClgH3gSn: 387.2199. Found: 387.2139.
Tributylstannyl-2-methyl-1(Z)-decene (93d): 0.85 g (38%); 1~ NMR
(CDC13) 6 0.88 (t, 12H, CH3, J = 6.6 Hz), 1.2-1.4 (m, 22H, CH2), 1.45- 1.6 (m, 8H,
CHZ), 1.78 (s, 3H, CH3), 2.23 (dt, 2H, C=CCH2, J = 6.7 Hz, 2.0 HZ), 5.45 (s, lH,
3 ~ ~ n - ~ = 75 Hz); 13c NMR 6 155.3 (C=CSnBug), 121.3 (C=CH), 42.2, 32.0, 29.5,
29.3, 29.2, 28.2, 27.3, 24.4, 22.7, 19.7, 13.6, 10.0; 119sn NMR = - 49.0; GCMS,
mle (relative intensity), 387 (M+ - 56, 100); Anal Calcd for ClgH3gSn: 387.2199.
Found: 387.2169.
b) Palladium-Catalyzed Cross-Coupling Reactions or i,2-
Dimetallo-1-alkenes, Preparation of 92c, 92e and 92f: To 0.74 g (1 mmol) of
Pd(PPh3)2C12 in 20 mL of THF were sequentially added DIBALH (2 mL, 2 mmol;
25OC, 30 min), cis 1,2-dimetallo-vinyl adduct (10 mmol, prepared in a separate flask as
described above) and the electrophilic coupling reagent (15.0 mmol). The reactions were
stirred at overnight at room temperature. Usual workup followed by silica gel
chromatography (n-hexanes as eluant) gave the bifunctionalized vinylstannanes. Cross-
coupled products derived from reaction of the vinylstannyl bonds were not detected.
For 92c: 2.1 g (89%).
1-Phenyl-3-tributylstannyl-2(Z)-undecene (92e): 2.17 g (84%); 1~
NMR (CDC13) 6 0.88 (t, 12H, CH3, J = 7.0 Hz), 1.2- 1.4 (m, 22H, CH2), 1.45- 1.6
(m, 8H, CH2), 2.25 (dt, 2H, C=CCH2, J = 7.0 Hz, 2.0 Hz), 3.4 (d, 2H,
C=CCH2Ph, J = 7.0 Hz), 6.2 (tt, lH, CzCCHCH2, J = 7.0 Hz, 2.0, 3~Sn-H = 140
Hz), 7.15-7.4 (m, 5H, Ph); l19sn NMR = - 50.1; GCIMS, rn/e (relative intensity), 463
(M+ - 56, 100).
1-Phenyl-2-tributylstannyl-1(Z)-decene (920: 2.0 g (79%); 1~ NMR
(CDC13) 6 0.88 (t, 9 H, CH3, J= 7.0 Hz), 0.9 (t, 3H, CH3, J = 7.0 Hz), 1.2-1.4 (m,
12H, CH2), 1.45-1.6 (m, 18H, CH2), 2.25 (dt, 2H, C=CCH2, J = 7.0 Hz, 1.5 Hz),
6.15 (t, lH, C=CH, J = 1.5 Hz, 3 ~ ~ n - ~ = 135 Hz), 7.1-7.4 (m, 5H, Ph); 119sn NMR
= - 47.0; GC/MS, m/e (relative intensity), 449 (M+ - 56, 100).
General Procedure for the Preparation of Disubstituted
Vinylstannanes (Scheme 40) 1-Nonyne (0.38 g, 3.0 mmol) in 5 mL of THF was
added to a solution of 94 at -65OC (5.8 mmol, vide supra) in 20 mL of THF. This was
followed by addition of CuBr (0.06 g, 0;3 mrnol) at which point the reaction turned
orange. After stirring for 3 h at -65OC CuBr (0.6 g, 3.0 rnmol) was added and the
reaction allowed to stir further for an hr. Ally1 bromide (0.8 mL, 9.0 mrnol) was then
added and the reaction was warmed to O•‹C (wine red in color). Oxidation with NaOH(15
mL) and H202 (5 mL), followed by column purification (hexanes as the eluant) resulted
in the desired products in tbe yields reported in Scheme 40.
In cases where PdO was the coupling agent, the initial reaction solution containing
the 12-dimetallo adduct was added via a canula to a solution of the electrophile (6.0
rnmol) in 5 mL of THF and PdO (0.15 rnmol) at room temperature. The palladium
catalyzed reactions were stirred further for overnight. All reactions were subjected to
normal workup followed by flash chromatography to give the indicated products whose
spectral data are reported below:
5-Tri butylstannyl-4(Z)-1-dodecadiene (96c): 1.16 g (89%); 1~ NMR
(CDC13) 6 0.88 (t, 12H, CH3, J = 7.0 Hz), 1.2-1.4 (m, 20H, CH2), 1.45- 1.6 (m, 8H,
CH2), 2.23 (dt, 2H, C=CCH2, J = 7.0 Hz, 1.5 Hz), 2.73 (ddt, 2H, C=CCH2C=C, J =
7.0 Hz, 2.0 Hz, 1.5 Hz), 4.99 (ddt, lH, HC=CHcis, J = 10.0 Hz, 2.0 Hz, 1.5 Hz),
5.02 (dq, 1 H, HC=CHtrans, J = 17.0 Hz, 2.0 Hz), 5.8 (ddt, 1 H, HC=CH2, J = 17.0
Hz, 10.0 HZ, 6.0 HZ), 5.96 (ddt, IH, C=CHCH2, J = 7.0 HZ, 2.0 HZ, 1.5 HZ, 3 ~ ~ n - H
= 138 Hz); 1 3 ~ NMR 6 156.0 (C=CSnBu3), 148.0 (C=CH), 139.5 (HC=CH2), 138.0
(C=CH2), 115.0 (C=CCH2C=C), 41.4, 32.0, 29.3, 29.2, 29.1, 27.4, 22.6, 13.6,
9.7; 119~11 NMR = - 52.1; GC/MS, m/e (relative intensity), 399 (M+ - 56, 100).
3-Keto-7-tributylstannyl-6(Z)-tetradecene (96d): 0.8 g (53%); 1~
NMR (CDC13) 6 0.88 (t, 12H, CH3, J = 7.U Hz), 1.02 (t, 3H, CH3), 1.2-1.4 (m,
20H, CH2), 1.45- 1.6 (m, 8H, CH2), 2.08 (q, 2H, CH2), 2.15 (t, 2H, CH2), 2.25 (dt,
2H, C=CCH2, J = 7.0 Hz, 2.0 Hz), 5.92 (tt, lH, C=CCHCH2, J = 7.0 Hz, 2.0,
3 ~ ~ n - ~ = 140 Hz); IR (cm-l) 2966.1726.1455; 119sn NMR = - 57.3; GC/MS, m/e
(relative intensity), 443 (M+ - 56, 100).
1-Phenyl-3-tributylstannyl-2(Z)-decene (96e): 1.18 g (78%); 1~ NMR
(CDC13) 6 0.88 (t, 12H, CH3, J = 7.0 Hz), 1.2- 1.4 (my 20H, CH2), 1.45- 1.6 (my 8H,
CH2), 2.25 (dt, 2H, C=CCH2, J = 7.0 Hz, 2.0 Hz), 3.4 (d, 2H, C=CCH2Ph, J = 7.0
Hz), 6.2 (tt, lHy C=CCHCH2, J = 7.0 Hz, 2.0, 3 ~ ~ n - ~ = 140 Hz), 7.15-7.4 (my 5H,
Ph); l 19sn NMR = - 50.1; GC/MS, m/e (relative intensity). 449 (M+ - 56, 100).
1-Phenyl-2-tributylstannyl-l(2)-nonene (960: 0.92g (63%); 1~ NMR
(CDC13) 6 0.88 (t, 9 H, CH3, J = 7.0 Hz), 0.9 (t, 3H, CH3, J = 7.0 Hz), 1.2-1.4 (my
12H, CH2), 1.45-1.6 (my 18H, CH2), 2.25 (dt, 2H, C=CCH2, J = 7.0 Hz, 1.5 Hz),
6.15 (t, lH, C=CHy J = 1.5 HZ, 3 ~ ~ n - H = 135 Hz), 7.1-7.4 (my 5H, Ph); l19sn NMR
= - 47.0; GC/MS, m/e (relative intensity), 435 (M+ - 56, 100).
Stannylalumination of Functionalized 1-Alkynes (Scheme 41)
Preparation of 6-Acetoxy-1-hexyne (98b): Pyridine (5.45 g, 0.069
mol), Ac20 (6.9 g, 0.068 mol) and 5-hexyn-1-01 (98a, 6.7 g, 0.069 mol) were stirred
together for 3 h at room temperature after which time the reaction was worked up in the
normal manner. Vacuum distAlation yielded 9.25 g (97%) of the acetate. (bp 61•‹C @ 10
mm Hg). GC analysis revealed i p ~ % ~ u; 971. !:I :.?X (CDC13) 6 1.62 (pentet, 2H,
CH2, J = 6.6 Hz), 1.77 (pentet, 2H, CH2, J = 6.6 Hz), 1.96 (t, lH, CCH, J = 1.9
Hz), 2.1 (s, 3H, COCH3), 2.22 (dt, 2H, CCCH2, J = 6.6 Hz, 1.9 Hz), 4.1 (t, 2H,
OCH2, J = 6.6 Hz); GC /MS, m/e (relative intensity), CI (isobutane) 141 (M+ +I, 100).
Preparation of 6-Tetrahydopyranyloxy-1-hexyne (98c): To 5-hexyne-
1-01 (9&, 5.0 g, 0.50 mol) and freshly distilled dihydropyran (10.5 g, 0.125 mol) were
added 4 drops o conc. He1 and the mixture stirred overnight at room temperature. Ether
(30 mL) was then added and the mixture shaken with 10% NaOH solution until neutral.
Usual workup followed by vacuum distillation yielded 8.4 g (91%) of the protected
alcohol (bp 108OC @ 17 mxn Hg). GC analysis revealed a purity of 98%. 1~ NMR
(CDC13), 6 1.2-1.8 (my 10Hy CH2), 1.96 (t, lH, CCH, J = 1.9 Hz), 2.22 (dt, 2H,
CCCH2, J = 6.6 Hz, 1.9 Hz), 3.38 (ddd, lH, OCH2CH2), 3.5 (dt, lH, CH2 on
OTHP), 3.75 (ddd, lH, OCH2CH2), 3.85 (dt, lH, CH2 on OTHP), 4.1 (tt, lH,
OCHO); GC IMS, d e (relative intensity), 182 (M+, 100).
Preparation of 1-Bromo-5-hexyne (98d): PBr3 (4.86 mL, 0.05 mmol)
was added dropwise to 5-hexyne-1-ol(98a, 13.72 g, 0.14 mol) in 50 rnL of anhydrous
Et20 at -5OC. The reaction was stirred for 2 h and then warmed to room temperature. It
was then quenched by pouring onto ice-cold NaHC03 solution. Usual workup followed
by distillation gave 19.8 g (88%) of the bromide (bp 68OC @ 21 mm Hg). GC analysis
revealed a purity of 92%. IH NMR (CDCl3) 6 1.73 (pentet, 2H, CH2, J = 6.6 Hz),
1.96 (t, lHy CCH, J = 1.9 Hz), 2.02 (pentet, 2H, CH2, J = 6.6 Hz), 2.30 (dt, 2H,
CCCH2, J = 6.6 Hz, 1.9 Hz), 3.5 (t, 2H, BrCH2, J = 6.6 Hz); GC/MS, m/e (relative
intensity), CI (isobutane) 163 (M+ + 2, 12.5), 161 (M+, 14.6).
Preparation of 6-Hydroxy-2-tributylstannyl-1-hexene (99a): To an
efficiently stirred solution of lithium diisopropyl amide (5.8 mmol) in 5 mL of THF was
added Bu3SnH (1.69 g, 5.8 mmol) in 5 mL THF at -30•‹C. After stirring for 30 min
EtpllCl(5.8 mL, 5.8 mmol) was added and the reaction was further stirred for 30 min.
5-Hexyne-1-ol,98a (0.29 g, 3.0 mmol), in 5 mL THF was added dropwise followed
by CuCN (0.026 g, 0.3 mmol). The reaction turned light mange at this point and was
stirred at -30•‹C for 3 h, and then overnight at room temperature. The usual workup,
followed by chromatography on silica gel (hexane:ethyl acetate, 96:4, as eluant) gave
0.97 g (84%) of a &&re of 99a and lOOa (2:l). IH NMR (CDC13) 6 0.88 (t, 14H,
CH3, J = 6.7 Hz), 1.32 (q, 18H, CH2, J = 6.7 Hz), 1.47 (m, 12H, CH2), 1.55 (m,
3H, OCH2CH2), 2.7-3.2 (m, 3H, C=CH2), 3.6 (rn, 3H, OCH2), 5.1 (dt, lH, C=CH,
Jz4.2 , 1.1, 3 ~ ~ n - ~ = 64 Hz), 5.7 (dt, lH, C=CH, J=4.2,2.9, 3 ~ ~ n - H = 140 HZ)),
5.9 (m, OSH, C=CH); l 3 ~ NMR 6 155.2 (C=CSnBu3), 149.2 (C=CSnBuj), 127.7
(C=CH2), 125.0 (C=CH2), 62.9 (C-OH), 62.8 (C-OH), 40.9, 37.5, 32.4, 32.3, 32.2,
29.3, 29.2, 29.1, 27.5,27.2,25.6, 25.1, 13.6, 9.6; GC/MS, m/e (relative intensity),
For 99a: 333 (M+-56, 100). For 100a: 333 (M+-56,40); Anal calcd for C14H2gOSn:
333.1241 Found: 333.1239.
Preparation of 6-Acetoxy-2-tributylstannyl-1-hexene (99b): 1-
Acetoxy-5-hexyne (98b, 0.39 g, 2.8 mmol), in 10 mL THF was added dropwise to an
efficiently stirred solution of Bu3SnAlEt2 (3.0 mmol, vide supra) followed by CuCN
(0.012 g, 0.14 mmol). The reaction was stirred at and further stirred for 2 h. Usual
workup, followed by silica gel chromatography (hexane:ethyl acetate, 955, as eluant)
gave 1.05 g (87.5%) of 99b in 99% purity as measured by GC analysis. 1~ NMR
(CDC1-3) 6 0.88 (t, 9H, CH3, J = 6.6 Hz), 1.32 (q, 12H, CH2, J = 6.6 Hz), 1.5 (m,
8H, CH2), 1.6 (t, 2H, CH20, J = 6.7 Hz), 2.1 (s, 3H, COCH3), 2.3 (tt, 2H,
C=CCH2, J = 6.6 HZ, 1.9 Hz), 4.1 (t, 2H, CH20, J = 6.6 Hz), 5.1 (dt, lH, C=CH, J
= 5.0, 1.2, 3 ~ ~ n - ~ = 63 HZ), 5.7 (dt, IH, C=CH, J = 5.0, 2.9, 3Jsn-H = 138 HZ);
l 3 ~ NMR 6 154.9 (C=CSnBu3), 125.2 (C=CH2), 64.4 (OCH2), 40.8, 29.1, 28.3,
27.3,25.9,20.9, 13.6.9.6; l 1 9 ~ n NMR = - 44.7; GC/MS, m/e (relative intensity), 375
(M+-56,61); Anal calcd for C16H3102Sn: 375.1346. Found: 375.1353.
Preparation of 6-Tetrahydropyranyloxy-2-tributylstannyl-1-hexene
(99c): A solution of Bu3SnAlEt2 (5.8 mmol) in 10 mL of THF was prepared as
described above. 6-Tetrahydropyranyloxy-1-hexyne (98c, 0.546 g, 3.0 mmol), in 5 mL
THF was added dropwise with stirring, followed by CuCN (0.02 g, 0.3 mmol ). The
reaction was stirred at -30•‹C for 3 h after which time it turned clear yellow. The reaction
was then warmed to room temperature and subjected to the usual workup. This was
followed by chromatography on silica gel (hexane:ethyl acetate, 99: 1, as eluant) to give
1.05 g (75%) of 95c in >99% purity. IH NMR (CDC13) 6 0.88 (t, 9H, CH3, J = 6.6
Hz), 1.32 (q, 12H, CH2, J = 6.6 Hz), 1.5 (m, 8H, CH2), 1.6 (t, 2H, CH20, J = 6.7
Hz), 2.3 (tt, 2H, C=CCH2, J = 6.6 Hz, 1.9 Hz), 3.38 (ddd, 1H, OCH2CH2), 3.5 (tt,
lH, CH2 on OTHP), 3.75 (ddd, lH, OCH~CHZ), 3.85 (tt, lH, CH2 on OTHP), 4.6
(tt, lH, OCHO), 5.1 (dt, 1Hy C=CH, J= 3.0, 1.2, 3 ~ ~ n - H = 64 HZ), 5.7 (dt, lH,
C=CHy J = 3.0, 1.2, 3 ~ ~ n - H = 140 Hz); l 3 ~ NMR 155.4 (C=CSnBu3), 124.8
(C=CH2), 98.7 (OCHO), 67.4 (OCH2)y 62.1 (OCH2), 41.1, 30.8, 29.5, 2961, 27.3,
26.3,25.5, 19.6, 13.6,9.6; GC/MS, m/e (relative intensity) 417 (M+-56,2.3); Anal
calcd for ClgH3702Sn: 417.1816. Found: 417.1808.
Preparation of 6-Bromo-2-tributylstannyl-1-hexene (99d). 1-Bromo-
5-hexyne (98d, 0.48gY 3 mmol) was added to a solution of Bu3SnAlEt2 (5.8 mmol) in
10 mL of THF followed by CuCN (0.02 g, 0.3 mmol). The reaction was stirred at -30•‹C
for 3 h, then warmed to room temperature. The usual workup, followed by
chromatography on silica gel (hexane, as eluant) gave 0.75 g (56%) of 99d and 30% of
tile cyciizeci pnxiu~t jiihi') presumably arising from the intramolecular cyclization of
the trans regioisomer (100d). IH NMR (CDC13) 6 0.88 (t, 9H, CH3, J = 6.6 HZ),
1.32 (q, 12H, CH2, J = 6.7 Hz), 1.40 (my 8H, CH2), 1.85 (t, 4H, CH2, J = 6.7 Hz),
2.3 (tt, 4H, C=CCH2, J = 6.6 Hz, 1.9 Hz), 3.41 (t, 4H, CH2Br, J = 6.6 Hz), 5.1 (dt,
lSH, C=CHy J = 3.0, 1.2, 3 ~ ~ n - ~ = 64 HZ), 5.7 (dt, lSH, C=CH, J = 3.0, 1.9,
3 ~ ~ n - ~ = 140 Hz); For cyclized product: IH NMR (CDC13) 6 5.9 (brs, OSH, C=CH,
2 ~ ~ n - ~ = 66 HZ); 13c NMR 6 155.4 (C=CSRBu3), 125.3 (C=CH2), 40.3 (CHzBr),
32.4, 30.6, 29.1,27.4,27;3, 13.6,9.6; GC/MS, m/e (relative intensity) For 99d: 395
(M+-56, 2); For 100d': 315 (M+-80,75); Anal calcd for C14H2gBrSn: 395.0397.
Found: 395.0398.
Preparation of 6-Cyano-2-tributylstannyl-l-hexene (99e): l-Cyano-
5-hexyne (98e, 0.24 g, 2.2 mmol), in 10 mL THF was added dropwise to an efficiently
stirred solution of Bu3SnBBN-OMe (3.0 rnmol, vide supra) followed by CuBr (0.04 g,
0.22 mmol). The reaction was stirred at -70•‹C for 3 h.CuBr 90.4 g, 2.2 mmol) was
then added followed by the addition of MeOH after 2 h of stirring. Usual workup,
followed by silica gel chromatography (hexane:ethyl acetate, 955, as eluant) gave 0.48 g
(81.5%) of 99e in 99% purity as measured by GC analysis. 1~ NMR (CDC13) 6 0.88
(t, 9H, CH3, J = 6.6 Hz), 1.32 (q, 12H, CH2, J = 6.6 Hz), 1.5 (m, 8H, CH2), 1.6 (t,
2H, CH2CN J = 6.7 HZ), 2.3 (a, 2H, C=CCH2, J = 6.6 Hz, 1.9 Hz), 5.1 (dt, lH,
C=CHy J = 5.0, 1.2, 3 ~ ~ n - ~ = 63 Hz), 5.7 (dt, IH, C=CH, J = 5.0, 2.9, ~ J S ~ - ~ =
138 Hz); l 3 ~ NMR 6 154.9 (C=CSnBu3), 125.2 (C=CH2), 64.4 (OCHZ), 40.8, 29.1,
28.3,27.3,25.9,20.9, 13.6,9.6; l19Sn NMR = - 44.7; GC/MS, m/e (relative
intensity), 342 (M+-56, 100).
General procedure for the preparation of disubstituted
functionalzed vinyl stannanes - Preparation of 9-Hydroxy-S-
tributylstannyl-4(E)-2-nonadiene (99f): To an efficiently stirred solution of
lithium diisopropyl amide (4.0 mmol) in 5 mL of THF was added Bu3SnH (1.0 mL, 3.8
mmol) in 5 mL THF at -30•‹C. After stirring for 30 min 9-BBNOMe (0.6 mL, 3.5
mmol) was added and the reaction was further stirred for 30 min. 9Hexyne-1-ol(98a)
(0.22 g, 2.2 mmol), in 5 mL THF was added dropwise followed by CuBr (0.04 g, 0.22
mmol). The reaction turned light orange at this point and was stirred at -70•‹C for 3 h, and
then added CuBr (0.4 g, 2.2 mmol). After i hr of stining ally1 bromide (0.23 g, 6.0
mmol) was added. The reaction was stirred overnight at room temperature. The usual
workup, followed by chromatography on silica gel (hexane:ethyl acetate, 6: 1, as eluant)
gave 0.53 g (57%) of 991 .IH NMR (CDC13) 6 0.88 (t, 9H, CH3, J = 6.7 Hz), 1.32
(q, 12H, CH2, J = 6.7 Hz), 1.47 (m, 8H, CH2), 1.55 (m, 2H, OCHzCHz), 2.32 (dt,
2H, CH2, J = 6.7 Hz, 1.5 Hz), 2.73 (ddt, 2H, C=CCH2C=C, J = 7.0 Hz, 2.0 Hz, 1.5
Hz), 3.6 (m, 2H, OCHz), 4.99 (ddt, lH, HC=CHcis, J = 10.0 Hz, 2.0 Hz, 1.5 Hz),
5.02 (dq, 1 H, HC=CHtrans, J = 17.0 Hz, 2.0 Hz), 5.8 (ddt, 1 H, HC=CH2, J = 17.0
Hz, 10.0 HZ, 6.0 HZ), 5.96 (ddt, lH, C=CHCH2, J = 7.0 Hz, 2.0 HZ, 1.5 HZ, 3~Sn-H
= 138 Hz); 13c NMR 6 155.2 (C=CSnBu3), 139.5 (HC=CH2), 127.7 (C=CH2),115.0
(C=CCH2C=C), 62.9 (C-OH), 40.9, 37.5, 32.4, 32.3, 32.2, 29.3, 29.2, 29.1, 27.5,
27.2, 25.6, 25.1, 13.6,9.6; GC/MS, m/e (relative intensity), For 99f: 373 (M+-56,
100).
Preparation of 9-Acetoxy-5-tributylstannyl-4(E)-2-nonadiene
(993): 1.26 g (73%) of 99g in 99% purity as measured by GC analysis. 1~ NMR
(CDC13) 6 0.88 (t, 9H, CH3, J = 6.6 Hz), 1.32 (q, 12H, CH2, J = 6.6 Hz), 1.5 (m,
8H, CH2), 1.6 (t, 2H, CH20, J = 6.7 Hz), 2.1 (s, 3H, COCH3), 2.3 (tt, 2H,
C=CCH2, J = 6.6 HZ, 1.9 Hz), 2.73 (ddt, 2H, C=CCH2C=C, J = 7.0 Hz, 2.0 Hz, 1.5
Hz), 4.1 (t, 2H, CH20, J = 6.6 Hz), 4.99 (ddt, lH, HC=CHcis, J= 10.0 Hz, 2.0 Hz,
1.5 Hz), 5.02 (dq, lH, HC=CHtrans, J = 17.0 Hz, 2.0 Hz), 5.7 (dt, lH, C=CH, J =
5.0,2.9, 3 ~ ~ n - ~ = 138 Hz); l 3 ~ NMR 6 154.9 (C=CSnBug), 125.2 (C=CH2), 115.0
(C=CCH2C=C), 64.4 (OCHZ), 40.8, 29.1, 28.3, 27.3, 25.9, 20.9, 13.6, 9.6; 1 1 9 ~ n
NMR = - 44.7; GC/MS, m/e (relative intensity), 415 (M+-56, 100).
Preparation of .9-Tetrahydropyranyloxy-S-tributylstannyl-4(E)-2-
nonadiene (99h): 0.81 g (53%) of 99h in 299% purity. IH NMR (CDC13) 6 0.88
(t, 9H, CH3, J = 6.6 Hz), 1.32 (q, 12H, CH2, J = 6.6 Hz), 1.5 (rn, 8H, CH2), 1.6 (t,
2Hy CH20y J = 6.7 HZ), 2.3 (tt, 2Hy C=CCH2, J = 6.6 HZ, 1.9 HZ), 2.73 (ddt, 2Hy
C=CCH2C=CY J = 7.0 Hz, 2.0 Hz, 1.5 Hz), 3.38 (ddd, lH, OCHZCH~), 3.5 (tt, lH,
CH2 on OTHP), 3.75 (ddd, lH, OCH~CHZ), 3.85 (tt, lH, CH2 on OTHP), 4.1 (t,
2H, CH20, J = 6.6 Hz), 4.6 (tt, 1 H, OCHO), 4.99 (ddt, 1 H, HC=CHcis, J = 10.0 Hz,
2.0 HZ, 1.5 Hz), 5.02 (dq, lH, HC=CHtrans, J = 17.0 Hz, 2.0 Hz), 5.7 (dt, lH,
C=CH, J = 3.0, 1.2, 3 ~ ~ n - H = 140 Hz); 1 3 ~ NMR 155.4 (C=CSnBu3), 124.8
(C=CH2), 1 15.0 (C=CCH2C=C), 98.7 (OCHO), 67.4 (OCH2), 62.1 (OCHz), 41 1 ,
30.8, 29.5, 29.1, 27.3, 26.3, 25.5, 19.6, 13.6, 9.6; GC/MS, m/e (relative intensity)
457 (M+-56, 52.3).
Preparation of 9-Bromo-S-tributylstannyi-4(E)-2-nonadiene (999.
0.78 g (53%); IH NMR (CDC13) 6 0.88 (t, 9H, CH3, J = 6.6 Hz), 1.32 (q, 12H,
CH2, J = 6.7 Hz), 1.40 (rn, 8H, CH2), 1.85 (t, 4H, CH2, J = 6.7 Hz), 2.3 (tt, 4H,
C=CCH2, J = 6.6 Hz, 1.9 Hz), 2.73 (ddt, 2H, C=CCH2C=C, J = 7.0 Hz, 2.0 Hz, 1.5
Hz), 3.41 (t, 4H, CH2Br, J = 6.6 Hz), 4.99 (ddt, lH, HC=CHcis, J = 10.0 Hz, 2.0
Hz, 1.5 HZ), 5.02 (dq, 1 H, HC=CHtrans, J = 17.0 HZ, 2.0 Hz), 5.7 (dt, 1 SHY C=CH,
J = 3.0,1.9, 3 ~ ~ n - ~ = 140 Hz); 13c NMR 6 155.4 (C=CSnBu3), 125.3 (C=CH2),
115.0 (C=CCH2C=C), 40.3 (CHzBr), 32.4, 30.6, 29.1, 27.4, 27.3, 13.6, 9.6;
GC/MS, m/e (relative intensity) 435 (M+-56,48).
Preparation sf 9-Cyano-5-tributylstannyl-4(E)-2-nonadiene (99j).
1.02 g (78%); IH NMR (CDC13) 6 0.88 (t, 9H, CH3, J = 6.6 Hz), 1.32 (q, 12H,
CH2, J = 6.7 Hz), 1.40 (m, 8H, CH2), 1.85 (t, 4H, CH2, J = 6.7 Hz), 2.3 (tt, 4H,
C=CCH2, J = 6.6 Hz, 1.9 Hz), 2.73 (ddt, 2H, C=CCH2C=C, J = 7.0 Hz, 2.0 Hz, 1.5
Hz), 3.41 (t, 4H, CH2Br, J= 6.6 Hz), 4.99 (ddt, lH, HC=CHcis, J = 10.0 Hz, 2.0
Hz, 1.5 Hz), 5.02 (dq, 1 H, HC=CHtrans, J = 17.0 Hz, 2.0 Hz), 5.7 (dt, 1 SH, C=CH,
J = 3.0, 1.9, 3 ~ ~ n - ~ = 140 HZ); 13c NMR 6 155.4 (C=CSnBu3), 125.3 (C=CH2),
115.0 (C=CCH2C=C), 40.3 (CHzCN), 32.4, 30.6, 29.1, 27.4, 27.3, 13.6, 9.6;
GC/MS, d e (relative intensity) 382 (M+-56, 100).
Preparation of Trisubstituted Alkenes (Schemes 42 and 43)
Preparation of 5-Iodo-4(Z)-1-tridecadiene (101a): To a solution of
92c (1.4 g, 3 mmol) in CH2C12 (10 mL) was added a solution of 12 in CH2C12
dropwise at -50•‹C until a faint coloration persisted The reaction was warmed to -20•‹C
and quenched with satd. NH4Cl. The vinyl iodide was extracted into pentane and-
combined extracts were washed with water and sodium thiosulphate and then dried over
potassium carbonate. silica gel chromatography with hexane as the eluant gave lOla in
72% (0.65 g) yield; IH NMR (CDC13) 6 0.87 (t, 3H, CH3, J = 7.1 Hz), 1.26-1.4 (m,
12H, CHz), 2.23 (dt, 2H, C=CCH2, J = 7.0 Hz, 2.0 Hz), 2.73 (dt, 2H,
C=CCH2C=C, J = 6.0 Hz, 2.0 Hz), 4.99 (dq, lH, C=CHcis, J = 10.0, 2.0 Hz), 5.02
(dq, lH, C=CHtrans, J = 17.0, 2.0 Hz), 5.8 (ddt, lH, CH2HC=CCH2, J = 17.0,
10.0, 6.0 Hz), 6.96 (tt, lH, C=CHy J = 6.0 Hz, 2.0 Hz); GC/MS, m/e (relative
intensity), 306 (M+, 40).
Preparation of lOlb and lOlc from 101a: n-BuLi (1.5 mL, 3.6 mmol)
was added to lOla (0.49 g, 1.6 mmol) at -78OC. After 0.5 h the reaction was quenched
with excess MegSiCl(101b) or 2 ~ 2 0 (101c) and slowly warmed to room temperature.
Usual workup followed by column chromatography with n-hexanes as eluant gave the
desired products in the yields listed below.
5-Trimethylsilyl-4(Z)-1-tridecadiene (101b): 0.3 g (78%); 1~ NMR
(CDC13) 6 0.15 (s, 9H, CH3Si), 0.87 (t, 3H, CH-3, J = 7.1 Hz), 1.26-1.4 (m, 12H,
CH2), 2.23 (dt, 2H, C=CCH2, J = 7.0 Hz, 2.0 Hz), 2.73 (dt, 2H, C=CCH2HC=C, J
= 6.0 Hz, 2.0 Hz), 4.99 (dq, lH, C=CHcis, J = 10.0, 2.0 Hz), 5.02 (dq, lH,
C=CHtrans, J = 17.0, 2.0 Hz), 5.8 (ddt, lHy CH2HC=CH2, J = 17.0, 10.0, 6.0 Hz),
5.96 (tt, 1Hy C=CH, J = 6.0 Hz, 2.0 Hz); 13c NMR 6 148.5 (C=CSiMe3), 139.3
(C=CH2), 137.5 (C=CH), 137.4 (C=CHCH2)y 114.8 (C=CHCH2CH=C), 38.4, 36.2,
32.0, 29.5, 29.4, 29.3, 22.7, 13.6,0.3; GC/MSy mle (relative intensity), 252 (M+,
100). Anal Calcd for C 16H32Si: 252.2273. Found: 252.228 1.
5-Deuterio-4(E)-1-tridecadiene (101c) 0.25 g (8740); IH NMR (CDC13)
6 0.87 (t, 3H, CH3, J = 7.1 Hz), 1.26- 1.4 (my 12H, CH2), 2.20 (dt, 2H, C=CCH2, J
= 7.0 Hz, 2.0 Hz), 2.73 (dt, 2H, C=CCH2HC=CY J = 6.0 Hz, 2.0 Hz), 5.0 (dq, lH,
C=CHcis, J = 10.0, 2.0 Hz), 5.02 (dq, 1Hy C=CHtrans, J = 17.0, 2.0 Hz), 5.40 (tt,
lH, C=CHy J = 6.0 Hz, 2.0 Hz), 5.8 (ddt, 1Hy CH2HC=CH2, J = 17.0, 10.0, 6.0
Hz); 3~ NMR 6 137.6 (HC=CH2), 13 1.9 (C=CD), 127.5 (C=CH), 1 14.8
(relative intensity), 181 (M+, 52). Anal Calcd for c ~ ~ H ~ ~ ~ H : 18 1.1941. Found:
181.1933.
lOlb was also prepared from 92c via transmetallation with n-BuLi. Thus, to a
solution of 92c (1.17 g, 2.5 mmol) in 5 mL of THF, was added n-BuLi (1.2 mL, 3
mmol) in 5 mL of TMEDA at -60•‹C. The reaction was stirred at this temperature for 2 h
and then warmed to room temperature. After stining further for 1 h, Me3SiC1(0.6 mL, 5
mmol) was added slowly at -78OC. The reaction was warmed to room temperature and
worked up as usual. silica gel chromatography with hexane as the eluant gave lOlb in
63% (0.39 g) yield.
General Procedure for Palladium Catalyzed Cross-Coupling Reactions:
Preparation of lOld and 101e: To a Schlenk tube at room temperature was
sequentially added Pd(Ph3P)4 (5 mole%) in 5 mL of benzene, benzylbromide (6 mmol)
or ally1 bromide (6 mmol) and vinylstannane 92d (2.36 g, 5.0 mmol). The reaction was
refluxed until palladium metal precipitated (24 h). The reaction was cooled to room
temperature and partitioned between Et20 (20 mL) and satd. KF (20 mL). After 0.5 h of
vigorous stirring, Bu3SnF was removed by filtration, the organic layer was separated,
washed with brine and dried. The solvent was evaporated and the crude mixture passed
through a silica gel column (hexanes:ethyl acetate, 99:1, as eluant) to give lOld or
10le .
1-Phenyl-2(Z)-(1-propeny1)-1-nonene (101d): IH NMR (CDC13) 6 0.9
(t, 3H, CH3, J = 7.0 Hz), 1.2-1.5 (m, 12H, CH2), 1.4 (d, 3H, CH3, J = 6.6 Hz), 2.1
(dt, 2H, C=CCH2, J = 7.0 Hz, 1.9 Hz), 3.3 (s, 2H, C=CCHzPh), 5.5 (qt, 1 H, C=CH,
J = 6.6, 1.9 Hz), 7.0-7.5 (m, 5H, Ph); 1 3 ~ NMR 6 149.6, 136.1, 129.0, 128,2,
126.0, 110.9, 65.8, 43.0, 35.5, 31.9, 29.4, 29.3, 29.2, 22.7, 22.6, 14.0; GCIMS, m/e
(relative intensity), 244 (M+, 12.3). Anal Calcd for C 1 gH28: 244.2 19 1. Found:
244.2187.
4(Z)-(1-Propeny1)-1-dodecene (101e): IH NMR (CDC13) 6 0.9 (t, 3H,
CH3, J = 7.0 Hz), 1.2-1.5 (m, 12H, CH2), 1.6 (d, 3H, CH3, J = 7.0 Hz), 1.95 (dt,
2H, C=CCH2, J = 7.0 Hz, 1.5 Hz), 2.8 (ddd, 2H, C=CCHzC=C, J = 7.0 Hz, 2.0 Hz,
1.5 Hz), 4.99 (ddd, lH, C=CHcis, J = 10.0 Hz, 2.0 Hz, 1.5 Hz ), 5.02 (dq, lH,
C=CHtrans, J = 17.0, 2.0 Hz), 5.3 (qt, lH, C=CH, J = 7.0 Hz, 1.5 Hz ), 5.75 (ddt,
lH, HC=CCH2, J = 17.0 Hz, 10.0 Hz, 7.0 Hz); GCIMS, m/e (relative intensity), 194
(M+, 100).
Preparation of (E)-1-Iodo-1-hexene (101f): This compound was
prepared by the known procedme.glb IH NMR (CDC13) 6 0.89 (t, 3H, CH3, J = 7.0
Hz), 1.26-1.4 (m, 4H, CH2), 1.98-2.0 (dt, 2H, J = 7.0 Hz, 1.9 Hz), 5.95 (dt, lH,
C=CHI, J = 15.0, 1.8 Hz), 6.5 (dt, lH, CH=CHI, J = 14.8,7 Hz). 1 3 ~ NMR 146.6,
73.9, 35.6, 30.5, 21.9, 13.7.
Preparation of 8-Methyl-S(E), 7(E)-hexadecadiene (101g): To a
solution of "Pd(Ph3P)f' (0.05 mnrz!) i,n, 5 ZL cf TEF, (,n=r.zrated in situ by the reaction
of DIBALH (0.1 mL, 0.1 mmol) and Pd(Ph3P)2C12 (0.037g, 0.05 mmol in THF), were
sequentially added 93d, (0.50 g, 1.2 mmol) and lOlf (0.21 1 g, 1.0 mmol) at room
temperature. The homogeneous reaction mixture turned black in 24 h. The reaction
mixture was diluted with ether and stinred with satd KF. After filtration of Bu3SnF, the
reaction was worked up as usual and the crude product pded by column
chromatography (hexane as eluant) to give 0.14 g (88%) of 101g. 1~ NMR (CDC13) 6
0.86 (t, 6H, CH3, J = 7.0 Hz), 1.02-1.4 (m, 16H, CH2), 1.9 (s, 3H), 2.3 (dt, 2H,
C=CCH2, J = 7.0 HZ, 1.6 Hz), 2.9 (dq, 2H, C=CCH2, J = 7.0, 1.6 Hz), 5.5 (ddt, lH,
HC=CH-C, J = 18.0 Hz, 10.0 HZ, 1.6 HZ), 5.8 (dt, lH, C=CH-HC=C, J = 10.0 Hz,
1.6 Hz), 6.0 (ddt, lH, CH=CH-C, J = 18.0 Hz, 10.0 Hz, 7.0 Hz); 1 3 ~ NMR 6 142.6
(C=CHCH=C), 1 32.7 (C=CHCH=C), 129.6 (C=CHCH=C), 123.5 (C=CHCH=C),
33.2, 32.7, 31.9, 29.9, 29.3,22.6, 17.6, 14.1; GC/MS, m/e (relative intensity) 280
(M+, 35.0). The spectral data matched those published for this compund.~l
Preparation of 1-Hexynyltri butylstannane (101h): To a solution of 1-
hexyne (0.82 g, 10.0 rnmol) in Et20 (15 mL) was added dropwise n-BuLi (3.8 mL,
10.0 mmol) at -30•‹C. After 30 min BqSnCl(3.25 g, 10.0 mmol) was added. The
reaction was warmed to mom temperature and then subjected to n d work-up. Bulb-
to-bulb distillation (bath temperature, 65OC @ 0.03 mm ~ g ) gave 3.5 g (95%); 1~ NMR
(CDC13) 6 0.86 (t, 9H, CH3, J = 7.0 HZ), 0.89 (t, 3H, CH3, J = 7.0 Hz), 1.2-1.4 (m,
12H, CH2), 1.4- 1.5 (m, 10H, CH2),-2.20 (t, 2H. CH2, J = 7.0 Hz); 1 1 9 ~ n = - 68.2;
GC/MS, m/e (relative intensity), 315 (M+ - 56, 100).
Preparation of 8-Methyl-7(E)-hexadecen-5-yne (101i): To a solution
of 93d (0.21 1 g, 1.0 rnmol, prepared as described earlier) and lOlh (0.44 g, 1.2
mmol) at room temperature was added "Pd(Ph3P)2" (0.05 mmol) in 5 mL of THF,
(generated in situ by the ?ZX",Z ~ v f 2 q i V - L c ; ~ ;f DIBALH and one equivalent of
Pd(Ph3P)zC12 in THF). The homogeneous reaction mixture turned black within an
hour. The black reaction mixture was added to 25 mL of water and this aqueous
mixture was extracted with ether (3 x 25 m.) which was back extracted with brine (1 x
25 mL) and dried over potassium carbonate. The dried extracts were Ntered through
alumina and concentrated under reduced pressure. The crude product was purZed by
column chromatography (hexane as eluant) to give 0.l55g (88%) of 101i. IH NMR
(CDC13) 6 0.87 (t, 6H, CH3, J = 6.7 Hz), 1.26- 1.4 (m, 12H, CH2), 1.54- 1-64 (m,
4H, CH2), 1.9 (s, 3H, CH2), 2.1 (dt, 2H, CCCH2, J = 6.7 Hz, 1.8 Hz), 2.23 (dt, 2H,
C=CCH2, J = 6.7 Hz, 1.9 Hz), 5.7 (t, lH, C=CH-CC, J = 1.9 Hz); 1% NMR 8
142.9 (C=CHCC), 120.1 (C=CH), 110.1. 88.6, 79.3, 39.1, 32.7, 32.5, 31.0, 22.1,
22.0, 19.0, 13.7, 13.5; GC/MS, m/e (relative intensity) 278 (M+, 23.0).
Synthesis of Square-Necked Grain beetle pheromone:
Synthesis of 104: CuCN (0.36 g, 4.0 mmol) was placed in a flask equipped with an
argon inlet. The flask was repeatedly (3x) evacuated (vacuum pump) and purged with
argon. THF (10 mL) was injected, the reaction was cooled to -50•‹C when MeLi (2.84
mL, 4.0 mrnol) was added slowly keeping the temperature below -50•‹C. The reaction
was stirred for 0.5 h and dimethylphenylsilyllithium in THF (4.2 mL, 4.0 rnrnol) added
dropwise. The resulting deep red solution was stirred for 0.5 h after which 1-butyne
(0.20 g, 4.0 mmol) was added dropwise. The reaction mixture was stirred for additional
2.0 h and then quenched with Me1 (2.0 mL) in HMPA (3.0 mL). The reaction mixture
was warmed gradually to room temperature. The usual workup followed by column
chromatography (nexanes) yleideci U. I L ~ g of 104 (90%) in > 95% purity as judged by
capillary g.c analysis. IR (NaCI) 1620, 1440, 1250, 1120,995 cm-l; 400 MHz IH
NMR (CDC13) 6 7.1-7.5 (m, 5H, Ph), 5.35 (dt, J = 18, 1.5 Hz, lH, C=CHSi), 2.7
(tdd, J = 7, 6, 1.5 Hz, 2H, allylic), 1.85 (s, 3H, C=CCH3), 0.95 (t, J = 7 Hz, 3H,
CH3), 0.3 (s, 6H, SiCH3); 1 3 ~ { l ~ ) (CDC13) 6 159.5 (C=CSi), 140.4 (C=CSi),
133.8 (ipso), 128.6 127.7, 118.8, 35.1, 22.0, 12.5, -0.9 (SiCH3); MS m/e (rel.
intensity) 204.3 (M+, 30), 189.2 (100), 135 (go), 121 (60); Anal. calc. C13H20Si
204.1337 found 204.1334.
Synthesis of 102: Synthesis of 102 was achieved by converting the vinyl silane
adduct to vinyl iodide and then cuprate followed by 1,4-addition of ethyl vinyl ketone by
known procedures. Reduction of the crude material followed by acetylation completed
the sequence. The following spectral data was obtained:
(E)-7-Met hyl-6-nonen-3-01: Conversion of vinylsilane to iodide (I2 in
CH2C12) followed by purification by flash chromatography resulted in the vinyl iodide.
Addition of 1.2 equiv. n-BuLi followed by 1 equiv. of CuBr and 1.2 equiv. of ethyl
vinyl ketone gave the vinyl ketone on workup. LAH reduction gave the alcohol. IR (cm-
l) 3360,2975, 1460,1120,970,850; 400 MHz H NMR (CDC13) 6 5.15 (dt, J = 18,
1.5 Hz, lH, C=CH), 3.5 (lH, CHOH), 2.10 (m, 2H, allylic), 2.7 (tdd, J = 7, 6, 1.5
Hz, 2H, allylic), 1.65 (s, 3H, C=CCH3), 1.4- 1.5 (m, 2H, CH2), 0.95 (t, J = 7 Hz,
3H, CH3); MS CI (isobutane), m/e (rel. intensity) 157 (M++ 1, 100); Anal. calc.
C10H200 C, 76.86; H, 12.86, found C, 76.74; H, 12.92.
(E)-3-Methyl-7-acetoxy-3-nonene (102): Conversion of the alcohol to the
acetate 102 with acetic anhydride in pyridine yielded >95% of 102 in > 95% purity (gc
analysis). IR (cm-l) 2975,1735,1460,1020,955; 400 MHz IH NMR (CDC13) 6 5.10
(dt, J = 18, 1.5 Hz, lH, C=CH), 4.9 (lH, CHOH,pentet, J=6 .8 Hz),2.0 (s, 3H,
COCH3), 2.13 (tdd, J = 7, 6, 1.5 Hz, 4H, allylic), 1.4-1.5 (m, 4H, CH2), 1 .O1 (t, 3H,
C=CCH3, J = 7 Hz), 0.85 (t, J = 7 Hz, 3H, CH3); MS CI (isobutane), m/e (rel.
intensity) 199 (M++ 1,5); Anal. calc. C12H2202 C, 72.68; H, 1 1.18, found C, 72.85;
H, 11.62.
Reaction of 1-Decynyldiethyl Aluminum with Tributyltin Hydride:
To a solution of 1 -decynyldiethyl aluminum (107) in THF (prepared from 1 -decyne
(0.138 g, 1.0 mmol) in 5 mL of THF, n-BuLi (0.40 mL, 1.04 mmol) and Et2AlCI (1.0
mL, 1.0 -01); O•‹C, 0.5 h), Bu3SnH (0.291 g, 1.0 mmol) was added dropwise and the
reaction mixture stirred overnight at O•‹C. Only 1-decyne and Bu3SnH were recovered.
Vinyl stannane products were not detected by gas chromatographic analysis after the
normal workup.
Reaction of Bu3SnAlEt2 and Tributylstannyl Hydride: To a THF
solution of B q s n A l E ~ (1.0 mmol) prepared by Method b was added Bu3SnH (0.291
g, 1.0 mmol) at O•‹C and the reaction mix& was stirred at this temperature. Only
Bu3SnH was obtained upon the usual workup. Formation of hexabutylditin was not
observed even after 24 h.
Reaction of Bu3SnAIEt2 with Tributylstannyl Hydride in the
Presence of Catalyst: Bu3SnAlEt2 (1.0 mmol) was prepared according to Method
b. Bu3SnH (0.291 g, 1.0 mmol) and CuCN (0.004 g, 0.05 rnmol) in 5 mL of THF
were added to this solution at O•‹C. After stining for 0.5 h, the reaction mixture was
quenched with 1N HCI and subjected to normal work-up. Hexabutylditin (0.04 g, 69%)
was obtained as the only product after bulb to bulb distillation.
Reaction of Tributyltin Hydride with CuCN: No reaction was observed
when Bu3SnH was reacted with CuCN in THF under argon at O•‹C for 12 h.
Reaction of Bu3SnAIEt2 with CuCN: CuCN (0.004 g, 0.05 mmol) was
added to a solution of Bu3snAlE~ (1.0 mmol, Method b) in in 5 mL of THF. The
solution immediately turned red. Workup after 30 min, yielded 69% of hexabutylditin.
Reaction of 1-Decynyldiethyl Aluminum with Bu3SnAIEt2:
Decynyldiethyl aluminum (107) was transferred via a canula to a THF solution of
Bu3SnAlEt2 (1.0 mmol) while maintaining the temperature at O•‹C. The reaction mixture
was stirred overnight at O•‹C after which it was subjected to normal workup to give 1-
decyne.
REFERENCES AND NOTES
(1) (a) Collman, J.; HegMus, L.S.; Norton, J.R.; Finke, R.GPrinciples and
Applications oflrgamtransition Metal Chemistry Univ. Science Books, Mill Valley,
CA 1987. (b) Kochi, J.K. Organometallic Mechanism and Catalysis Academic Press,
New York 1987. (c) Negishi, Ei-ichi. Current Trends in Organic Synthesis
Pergarnon Press, Oxford 1983. (d) Davies, S.G. Organotransition Metal Chemistry.
Applications to Organic Synthesis Pergarnon Press, Oxford 1982.
(2) (a) Lipshutz, B.H. Synthesis 1987, 325. (b) Foulon, J.P.; Bourgain-Commercon,
M.; Normant, J.F. Tetrahedron 1986,42, 1389. (c) Taylor, R.J.K. Synthesis 1985,
364. ( d ) Lipshutz, B.H.; Wilhelrn, R.S.; Kozlowski, J.A. Tetrahedron 1984,40,
5005. (e) Normant, J.F.; Alexakis, A. Synthesis 1981, 841. ( f ) Posner, G.H. An
Introduction to Synthesis Using Organocopper Reagents Wiley Interscience, New
York 1980. (g) Normant, J.F. Pure Appl. Chem. 1978,50, 709.
(3) Gilman, H.; Jones, R.G.; Woods, L.A. J. Org. Chem. 1952,17, 1630.
(4) (a) Bertz, S.H.; Gibson, C.P.; Dabbagh, G. Tetrahedron Lett. 1987, 4251.(b)
Lipshutz, B.H.; Whitney, S.; Kozlowski, J.A.; Breneman, C.M. Tetrahedron Lett.
1986,4273. (c) Bertz, S.H. Tetrahedron Lett. 1980, 3 151. (d) Bertz, S.H. J. Org.
Chem. 1979,44,4967. (e) Ashby, E.C.; Noding, S.A. J. Org. Chem. 1979,44,
4371. (f') Ashby, E.C.; Lin, J.J. J. Org. Chem. 1977,44, 1099.
(5) (a) Lipshutz, B.H.; Ellsworth, E.L.; Behling, J.A.; Campbell, A.L.Tetrahedron Lett.
1988, 893. (b) Bertz, S.H.; Gibson, C.P.; Dabbagh, G. Organometallics 1988, 7,
227. (c) Knochel, P.; Yeh, M.C.P.; Berk, S.C.; Talbert, J. J. Org. Chem. 1988,53,
2392. (d) Lipshutz, B.H.; Parker, D.A.; Nguyen, S.L.; McCarthy, K.E.; Barton, J.;
Whitney, S.; Kotsuki, H. Tetrahedron 1986,42,2873. (e) Ender, E. Tetrahedron
1984,40, 641. ( f ) Macdonald, T.L.; Narayanan, B.A.; O'Dell, D.E. J. Org. Chem.
1981,46, 1504. ( g ) Posner, G.H. Organic React. 1975,22, 253.
(6) (a) Lipshutz, B.H.; Ellsworth, E.L.; Siahaan, T.J. J. Am. Chem. Soc. 1988,110,
4834. (b) Lindstedt, E-L.; Nilsson, M.; Olsson, T. J. Organornet. Chem. 1987,334,
255. (c) Yamamoto, Y. Angew. Chem. Int. Ed., Engl. 1986,25, 947. (d) Boeckman,
R.K.; Batra, T.E. J. Org. Chem. 1985,50, 3421. (e) Salomon, R.G. Acc. Chem. Res.
1985,18, 294. ( f ) Corey, E.J.; Boaz, N.W. Tetrahedron Lett. 1985, 6019. (g) Corey,
E.J.; Kyler, K.; Raju, N. Tetrahedron Lett. 1984, 51 15. (h) Lipshutz, B.H.; Parker,
D.A.; Kozlowski, J.A.; Nguyen, S.L. Tetrahedron Lett. 1984,5959. (i) Ghribi, A.;
Alexakis, A.; Normant, J.F. Tetrahedron Lett. 1984, 3083. (j) Nakarnura, E.;
Kuwajima, U. J. Am. Chem. Soc. 1984,106, 3368. (k) Lindell, S.D.; Elliot, J.D.;
Johnson, W.S. Tetrahedron Lett. 1984, 3947. (1) Lipshutz, B.H.; Kozlowski, J.A.;
Wilhelm, R.S. J. Org. Chem. 1983,48, 546. (m) Oppolzer, W.; Moretti, R.; Godel,
T.; Meumier, A.; Loher, H. Tetrahedron Lett. 1983,4971. (n) Yamamoto, Y.;
Yamamoto, S.; Yatagai, H.; Ishihara, Y.; Maruyama, K. J. Org. Chem. 1982,47, 119.
(0) Drian, C. Le.; Greene, A.E. J. Am. Chem. Soc. 1982,104, 5473. (p) Wasserman,
H.H.; Gambale, R.J.; Pulwer, M.J. Tetrahedron 1981,37, 4059.
(7) Lipshutz, B.H.; Kozlowski, J.A.; Breneman, C.MJ. Am. Chem. Soc. 1985,107,
3 197.
(8) (a) Ashby, E.J.; Watkins, J.J. J. Chem. Soc., Chem. Commun. 1976,784. (b)
Ashby, E.C.; Watkins, J.J. J. Am. Chem. Soc. 1977,99, 5312.
(9) (a) Hogan, R.J.; Scherr, P.A.; Weibel, A.T.; Oliver, J.P. J. Organomet. Chem.
1975,85, 265. (b) Scherr, P.A.; Hogan, R.J.; Weibel, A.T.; Oliver, J.P. J. Am.
Chem. Soc. 1974, 96, 6055. (c) Dessy, R.E.; Kaplan, F.; Coe, G.R.; Salinger, R.M.
J. Am. Chem. Soc. 1963,85, 1191. (d) Heinzer, J.; Oth, J.F.M.; Seebach, D. Helvic.
Chim. Acta. 1985,68, 1848.
(10) Clive, D.L.; Farina, V.; Beaulieu, P.L. J. Org. Chem. 1982,47,2572.
(1 1) House, H.O.; Koepsell, D.G.; Campbell, W.J. J. Org. Chem., 1972,37, 1003.
(12) (a) Bertz, S.H.; Dabbagh, G. J. Am. Chem. Soc. 1988,110, 3668. (b) Bertz,
S.H.; Dabbagh, G. Tetrahedron, Symposia-in-Print on Organocopper Chemistry 1989,
45,425. (c) Hallnemo, G.; Ullenius, C. Tetrahedron, 1983,39, 1621.
(13) (a) Corey, E.J.; Beames, D. J. Am. Chem. Soc. 1972,94, 7210. (b) Corey, E.J.;
Floyd, D.M.; Lipshutz, B.H. J. Org. Chem. 1978,43, 3418. (c) Enda, J.; Matsutani,
T.; Kuwajima, LTetrahedron Lett. 1984,5307. (d) Ledlie, D.B.; Miller, G. J. Org.
Chem. 1979,44, 1006.
(14) (a) Posner, G.H.; Whitten, C.E.; Sterling, J.J. J. Am. Chem. Soc. 1973,95,
7788. (b) Luong-Thi, N.T.; Riviere, H. Tetrahedron Lett. 1970, 1583. (c) Lipshutz,
B.H.; Kozlowski, LA.; Parker, D.A.; Nguyen, S.L.; McCarthy, K.E. J. Organomet.
Chem. 1985,285,437. (d) Malmberg, H.; Nilsson, M.; Ullenius, C. Tetrahedron lett.
1982, 3823.
(15) (a) Bertz, S.H.; Dabbagh, G.; Villacorta, G.M. J. Am. Chem. Soc. 1982,104,
5824. (b) Bertz, S.H.; Dabbagh, G. J. Chem. Soc., Chem. Cornmun. 1982, 1030. (c)
Schwartz, R.H.; Filippo, S.J. J. Org. Chem. 1979,44, 2705.
(16) (a) Lipshutz, B.H.; Kozlowski, J.A. J . Org. Chem. 1984,49, 1147. (b) Marino,
J.P.; Femadez de la Pradilla, R.; Laborde, E. J. Org. Chem. 1984,49, 5279. (c)
Hamon, L.; Levisalles, J. J. Organomet. Chem. 1983,25I, 133. (d) Gorlier, J.P.;
Hamon, L.; Levisalles, J.; Wagnon, J. J. Chem. Soc., Chem. Comrnun. 1973, 88. (e)
Lipshutz, B.H.; Kozlowski, J.A.; Wilhelm, R.S. J. Am. Chem. Soc. 1982,104,2305.
(f') Corey, E.3.; Pan, B.C.; Hua, D.H.; Deardorff, D.R. J. Am. Chem. Soc. 1982,
104, 6816. (g) Corey, E.J.; Hua, D.H.; Pan, B.C.; Seitz, S.P. J. Am. Chem. Soc.
1982,104, 68 18. (h) Lipshutz, B.H.; Wilhelm, R.S.; Floyd, D.M. J. Am. Chem. Soc.
1981,103, 7672. (i) Comu, R.J.P.; Guerin, C.; M'Boula, J. Tetrahedron Lett. 1981,
2985. (i) Acker, R.D. Tetrahedron Lett. 1978,2399. (k) Acker, R.D. Tetrahedron Lett.
1977,3402. (1) Hamon, L.; Levisalles, J. Tetrahedron, Symposia-in-Print on
Organocopper Chemistry 1989,45,489.
(17) (a) Lipshutz, B.H.; Kozlowski, J.A.; Wilhelm, R.S. J. Org. Chem. 1984,49,
3943. (b) Lipshutz, B.H.; Kozlowski, J.A.; Breneman, C.M. Tetrahedron Lett. 1985,
5911.
(1 8) Massey, A.G. Comprehensive Inorganic Chemistry, Pergamon Press, Oxford 3,
1976.
(19) Cotton, F.A.; Wilkinson, G. Advanced Inorganic Chemistry, Wiley Interscience,
New York 1972.
(20) (a) Hope, H.; Oram, D.; Power, P.P. J . Am. Chem. Soc. 1984,106, 1149. (b)
Edwards, P.G.; Gellert, R.W.; Marks, M.W.; Bau, R. J. Am. Chem. Soc. 1982,104,
2072. (c ) Khan, S.I.; Edwards, P.G.; Xuan, H.S.; Bau, R. J: Am. Chem. Soc. 1985,
107, 1682.
(21) (a) van Koten G.; Jastrzebski, J.T.B.H.; Muller, F.; Stam, C.H. J. Am. Chem.
Soc. 1985,107, 697. (b) van Koten G.; Jastrzebski, J.T.B.H.; Stam, C.H.; Brevard,
C. Biological and Inorganic Copper Chemistry, K.D. Karlin and Zubieta, J, Eds.,
Adenine Press, 1985, 267-285.
(22) van Koten G.; Jastrzebski, J.T.B.H.; Noltes, J.G. J. .Organornet. Chem. 1977,
140, C23.
(23) (a) van Koten G.; Jastrzebski, J.T.B.H.; Stam, C.H.; Niemann, N.C. J. Am.
Chem. Soc. 1984,106, 1880. (b) van Koten G.; Jastrzebski, J.T.B.H.; Noltes, J.G. J.
Am: Chem. Soc. 1979,101, 6593. (c) Hoedt, R.W.M.T.; van Koten, G.; Noltes, J.G.
J. Organornet. Chem. 1979,179,227. (d) van Koten, G.; Noltes, J.G. J. Organornet.
Chem. 1979,174, 367.
(24) Dempsey, D.F.; Girolami, G.S. Organometallics. 1988, 7, 1208.
(25) Stewart, K.R.; Lever, J.R.; Whangbo, M.-H. J. Org. Chem. 1982,47, 1472.
(26) (a) Eaborn, C.; Hitcock, P.B.; Smith, J.D.; Sullivan, A.C. J. Organornet. Chem.
1984, 263, C23. (b) Hope, H.; Olmstead, M.M.; Power, P.P.; Sandell, J.; Xu, X. J.
Am. Chem. Soc. 1985,107, 4337. (c) Martin, S.F.; Fishpaugh, J.R.; Power, J.M.;
Giolando, D.M.; Jones, R.A.; Nunn, C.M.; Cowley, A.H. J. Am. Chem. Soc. 1988,
110,7226. (d) Cowley, A.H.; Giolando, D.M.; Jones, R.A.; Nunn, C.M.; (e) Power,
J.M. J. Chem. Soc., Chem. Cornmun. 1988, 208. (f) van Koten G.; Jastrzebski,
J.T.B.H.; Noltes, J.G. Tetrahedron, Symposia-in-Print on Organocopper Chemistry
(27) (a) Fleming, 1.; Newton, T.W. J. Chem. Soc., Perkin Trans. I 1984, 1805. (b)
Fleming, I.; Newton, T.W.; Roessler, F. J. Chem. Soc., Perkin Trans. I 1981, 2527.
(28) (a) Cox, S.D.; Wiidl, F. Organometallics 1983,2, 184. (b) Piers, E.; Chong, J.M.
J. Org. Chem. 1982,47, 1602. (c) Westmijie, H.; Ruitenberg, K.; Meijer, J.; Vermeer,
P. Tetrahedron Lett. 1983, 2797. (d) Piers, E.; Chong, M.J.; Morton, H.E.
Tetrahedron Lett. 1981,4905. (e) Piers, E.; Morton, H.E. J. Org. Chem. 1980,45,
4263. ( 0 Piers, E.; Morton, H.E. J. Org. Chem. 1979,46, 3437. (g) Piers, E.;
Morton, H.E. J. Chem. Soc., Chem. Commun. 1978, 1033. (h) Hudec, J. J. Chem.
Soc., Perkin Trans I 1975, 1020.
(29) (a) Taddei, M.; Mann, A. Tetrahedron Lett. 1986,2913. (b) Fleming, I.; Rowley,
M.Tetrahedron Lett. 1986, 5417. (c) Fleming, I.; Sarkar, A.K. J. Chi,n. So;., C h ' .
Comrnun. 1986, 1199. (d) Fleming, 1.; Tarrett, N.K. J. Organomet. Chem. 1985,
264, 99. (e) Fleming, I.; Henning, R.; Plaut, H. J. Chem. Soc., Chem. Commun.
1984, 29. (0 Fleming, I.; Newton, T.W. J. Chem. Soc., Perkin Trans. I 1984, 119.
(g) Fleming, I.; Tarret, N.K. Tetrahedron Lett. 1983,4151. (h) Fleming, 1.; Marchi, D.
Synthesis 1981,560. (i) Piers, E.; Tse, H.L. Tetrahedron Lett. 1984, 3155. (i) Piers,
E.; Karunaratne, V. J. Chem. Soc., Chem. Commun. 1983, 935.
(30) (a) Chen, H-M.; Oliver, J.P. J. Organornet. Chem. 1986, 316,255. (b) Fleming,
I.; Taddei, M. Synthesis 1985, 898. (c) Fleming, 1.; Taddei, M. Synthesis 1985, 899.
(d) Chow, H-F.; Fleming, I. J. Chem. Soc., Perkin Trans. I 1984, 1815. (e) Piers, E.;
Chong, J.M. Can. J. Chem. 1988,66, 1425. ( f ) Piers, E.; Morton, H.E.; Chong, M.J.
Can. J . Chem. 1987,65, 78. (g) Piers, E.; Chong, M.J. J. Chem. Soc., Chem.
Comrnun. 1983,934. (h) Piers, E.; Chong, M.J. Tetrahedron, Symposia-in-Print on
Organocopper Chemistry, 1989,45,363.
(31) (a) Fleming, I.; Roessler, F. J. Chem. Soc., Chem. Commun. 1980, 276. (b)
Fleming, I.; Rowley, M.; Cuadrado, P.; Gonzalez-Nogal, A-M.; Pulido, F-J.
Tetrahedron, Symposia-in-Print on Organocopper Chemistry, 1989,45,413. (c)
Cuadrado, P.; Gonzalez-Nogal, A-M.; Pulido, F-J.; Fleming, I. Tetrahedron Lett.
1988, 1825. (d) Morizawa, Y.; Oda, H.; Oshima, K.; Nozaki, H. Tetrahedron Lett.
1984, 1163.
(32) (a) Fleming, I.; Pulido, F-J. J. Chem. Soc., Chem. Commun. 1986, 1010. (b)
Fleming, I.; Waterson, D. J. Chem. Soc., Perkin Trans. I 1984, 1809. (c) Bernhard,
W.; Fleming, 1.; Waterson, D. J. Chem. Soc., Chem. Commun. 1984, 28. (d) Ager,
D.J.; Fleming, I.; Patel, S.K. J. Chem. Soc., Perkin Trans. i i%i, 23iCI. (ej Fieming,
I.; Percival, A. J. Chem. Soc., Chem. Commun. 1978, 278. ( f ) Piers, E.; Lu, Y.F. J.
Org. Chem. 1988,53,926. (g) Piers, E.; Chong, M.J.; Keay, B.A. Tetrahedron Lett.
1985, 6265. (h) Piers, E.; Karunaratne, V. J. Org. Chem. 1983,48, 1774. (i) Seitz,
D.E.; Lee, S-H. Tetrahedron Lett. 1981,4909. (j) Seebach, D. Angao. Chem. Int.
Ed., Engl. 1979,8, 239. (k) Seebach, D. Amberg, W. Angao. Chem. Int. Ed., Engl.
1988,27, 171 8. (1) Still, C.W. J. Org. Chem. 1976,41, 3063.
(33) (a) George, M.V.; Paterson, D.J.; Gilman, H. J. Am. Chem. Soc. 1960,82, 403.
(b) Gilman, H.; Shiina, K; Aoki, D.; Gaj, B.J.; Wittenberg, D.; Breiman, T. J .
Organornet. Chem. 1968,13,323.
(34) (a) The infrared spectra were recorded on solutions wanning to room temperature.
No visible decomposition was observed within 0.5 h for silylcuprates. (b) All solutions
studied contained LiCl which solubilizes CuCN. The CuCN.2LiCl complex in THF
exhibits an absorbtion at VCN = 2148 cm-1. (c) The stannylcuprates turned black while
the IR were being run.
(35) Nakamoto, K Infrared and Raman Spectra of Inorganic and Coordination
compounds Wiley Interscience, New York 1986.
(36) Equilibrium constant was estimated using the expressions:
k3 - [ (PhMe$3)3CuLi2 ] [ LiCN ] K-2 = - = 3 4 f 10
(37) (a) Mann, B.E.; Taylor, B.F. 1 3 ~ NMR Data for Organometallic Compounds
Academic Press, New York, 1981, p.6. (b) Hanis, R.K. Nuclear Magnetic Resonance
Spectroscopy Pitman, London, 1983, p. 192.
(38) (a) Brevard, C.; Granger, P. Handbook of High Resolution Mdtinuclear NMR
Wiley Interscience, New York 1981. (b) 2 9 ~ i has natural abundance of 4.746, NMR
receptivity of 2.09 with respect to 1% and a gyromagnetic ratio of -5.3 146. (c)
Coleman, B. NMR of Newly Accessible Nuclei Academic Press, New York, 1983,2,
p. 197. (d) Marsmann, H. NMR: Basic Principles and Progress. 1981,17,65. (e)
Schraml, J.; Bellama, J.H. Detemnnrnation of Organic Structures by Physical Methods
Academic Press, New York, 1976,6.
(39) (a) Bodner, G.M.; May, M.P.; McKinney, L.E. Inorg. Chem. 1980,19, 195 1.
(b) For similar observations see Krentz, R.; Pomeroy, R. Inorg. Chem. 1985,24,
2976.
(40) (a) Buncel, E.; Venkatachalan, T.K.; Eliasson, B.; Edlund, U. J. Am. Chem. Soc.
1985,107, 303. (b) Olah, G.A.; Huandi, R.J. J. Am. Chem. Soc. 1980,102, 6989.
(41) (a) 7 ~ i has nanual abundance of 92.5890, NMR receptivity compared to 13c of
1540 and a gyromagnetic ratio of 10.396. (b) A minor signal at -2.05 ppm is also present
which we attribute to cleavage of THF by PhMe2SiLi. (c) Hartwell, G.E.; Allerhand, A.
J. Am. Chem. Soc. 1971,93, 4415. (2) Czx, EX.; Terr;., Y.X. J . Magnetic
Resonance 1974,14, 317. (e) Giinther, H.; Moskau, D.; Bast, P.; Schmalz, D. Angew.
Chem. Int. Ed., Engl. 1987,26, 1212.
(42) Gilman, H.; Schulze, F.J. J. Am. Chem. Soc. 1925,47, 2002.
(43) (a) Seitz, L.M.; Little, B.F. J. Organomet. Chem. 1969,18, 227. (b) Seitz, L.M.;
Brown, T.L. J. Am. Chem. Soc. 1966,88, 4140.
(44) Gay, I.D. Personal cornrnunication..
(45) (a) Still, C.W.; Macdonald, T.L. Tetrahedron Lett. 1976,2659. (b) Posner, G.;
Lentz, C.M. J. Am. Chem. Soc. 1979,101,934.
(46) (a) Still, C.W. J. Am. Chem. Soc. 1977,99, 4836. (b) Tarnborski, C.; Ford,
F.E.; Soloski, E.J. J. Org. Chem. 1963,28, 181. (c) Oliver, J.P.; Weibel, A.T. J
Organomet. Chem. 1974,82,281. (d) Wells, W.L.; Brown, T.L. J. Organomet.
Chem. 1968,4, 271.
(47) (a) McFarlane, W.; White, R.F.M. Techniques of High Resolution Multinuclear
NMR Spectroscopy Butterworth, London, 1972. 19sn has natural abundance of
8.588, NMR receptivity of 25.2 with respect to l 3 ~ and a gyromagnetic ratio of
-9.9756.
(48) (a) Brown, T.L.; Morgan, G.L. Inorg. Chem. 1963,2, 736. (b) Harrison, P.G.;
Ulrich, S.E.; Zuckerman, J.J. J. Am. Chem. Soc. 1971,93, 5398. (c ) Kaesz, H.D.;
Holmes, J.R. J. Am. Chem. Soc. 1961,83, 3903.
(49) (a) Kennedy, J.D.; McFarlane, W. J. Chem. Soc. Chem. Commun. 1974,983.
(b) Biffar, W.; Noth, N.; Pommerening, H.; Schwerthoffer, R.; Storch, W.;
Wrackmeyer, B . Chem. Ber. 198 1,114,49.
(50) (a) Lipshua, B.H.; Ellsworth, E.L.; Dimock, S.H.; Reuter, D.C. in 72"d Canadian
Chemical Conference and Exhibition, Victoria, BC., June 1989. (b) Lipshutz, B.H.;
Ellsworth, E.L.; Dimock, S.H.; Reuter, D.C. Tetrahedron Lett., 1989,2065.
(5 1) This is based on the observation that reactions of "R3SnCuW as well as "R3SiCu"
can be conducted in MeOH.
(52) Krauss, S.R.; Smith, S.G. J . Am. Chem. Soc. 1981,103, 141.
(53) (a) Corey, E.J.; Boaz, N. Tetrahedron Lett. 1985,26, 6015. (b) Carey, E.J.;
Boaz, N.Tetrahedron Lett. 1984,25,3063.
(54) (a) House, H.O. Accounts Chem. Res. 1976,9, 59. (b) House, H.O.; Umen,
M.J. J. Org. Chem. 1973,38, 3893.
(55) Smith, R.A.J.; Hannah, D.J. Tetrahedron 1979,53, 1183.
(56) (a) john so^, C.R.; Dutra, G.A. J. Am. Chem. Soc. 1973,95, 7777. (b) Johnson,
C.R.; Dutra, G.A. ,'. .A,;;;. ,"lici~. Six. 1!?73,95, 7783. (c) Casey, C.P.; Cesa, M.C. J.
Am. Chem. Soc. 1979,101, 4236.
(57) Berlan, J.; Koosha, K.; Battioni, J.P. Bull. Soc. Chim. France 11 1978, 575.
(58) Berlan, J.; Battioni, J.P.; Koosha, K. Bull. Soc. Chim. France 11 1979, 183 and
references cited therein.
(59) Christenson, B.; Olsson, T,; Ullenius, C. Tetrahedron, Symposia-in-Print on
Organocopper Chemistry 1989,45, 523.
(60) Christenson, B.; Ullenius, C. Pure and Appl. Chem. 1988,60,57.
(61) Lindstedt, E-L.; Nilsson, M.; Olsson, T. J. Organornet. Chem. 1987,334,255.
(62) (a) Jolly, P.W.; Mynott, R. A h . Organometal. Chem. 1981,19, 257.
(63) Salomon, R.G.; Kochi, J.K. J. Organomet. Chem. 1974,64, 135. .
(64) Lipshutz, B.H.; Ellsworth, E.L.; Siahaan, T.J. J. Am. Chem. Soc. 1989,111,
1351.
(65) Lipshutz, B.H.; Ellsworth, E.L.; Siahaan, T.J.; Shirazi, A. Tetrahedron Len.
1988, 6677.
(66) (a) Miyaura, N.; Yamah, K.; Suzuki, A. Tetrahedron Lett. 1979,3437. (b)
Miyaurz :.'.; SiiziZ, ,*,. ,'. CAc;;;. Soc., Chem. Commun. 1979, 866. (c) Zweifel, G.;
Buckland, S.J. J. Am. Chem. Soc. 1977,99, 3 184. (d) Brown, H.C. Organic
Synthesis via Organoboranes, Wiley-Interscience: New York 1975.
(67) (a) Okukado, N.; VanHorn, D.E.; Klinia, W.L.; Negishi, E. Tetrahedron Lett.
1978, 1027. (b) Zweifel, G.; Whitney, C.C. J. Am. Chem. Soc. 1967,89, 2753. (c)
Miller, J.A.; Zweifel, G. J. Am. Chem. Soc. 1983,105, 1383. (d) Negishi, E.
Aspects and Mechanism of Organometallic Chemistry, J.H. Brewster, Ed.; Plenum:
New York, 1978.
(68) (a) Negishi, E.; Valente, L.F.; Kobayashi, M. J. Am. Chem. Soc. 1980,102,
3298. (b) Negishi, E.; Takahashi, T. Aldrichimica Acta 1985,18,31.
(69) Negishi, E. Pure and Applied Chemistry 1981,53,2333.
(70) (a) Hibino, J-I.; Matsubara, S.; Morizawa, Y.; Oshima, K.; Nozaki, H.
Tetrahedron Lett. 1984,215 1. (b) Matsubara, S.; Hibino, J-I.; Morizawa, Y.; Oshima,
K.; Nozaki, H. J. Organometallic Chem. 1985,285, 163. (c) Nonaka, T.; Okuda, Y.;
Matsubara, S.; Oshirna, K.; Utimoto, K.; Nozaki, H. J. Org. Chem. 1986,51,47 16.
(71) (a) Hibino, J-I.; Nakatsukasa, S.; Fugami, K.; Matsubara, S.; Oshima, K.; Nozaki,
H. J. Am. Chem. Soc. 1985, I07, 6416.
(72) (a) Nozaki, K.; Wakamatsu, K.; Nonaka, T.; Tiickrnantel, W.; Oshima, K.
Utirnoto, K. Tetrahedron Lett. 1986,2007. (b) Bihlmayer, C.; Kerschl, S.;
Wrackmeyer, B. 2. Naturforsch. 1987,42b, 715. (c) Chu, K-H.; Wang, K.K. J. Org.
cnem. 1~66, Ti, 767.
(73) (a) Mitchell, T.N.; Killing, H.; Dicke, R.; Wickenkamp, R. J. Chem. Soc., Chem.
Commun. 1985, 354. (b) Chenard, B.L.; Van Zyl, C.M. J. Org. Chem. 1986,5I,
3561. (c) Chenard, B.L.; Van Zyl, C.M.; Sanderson, D.R. Tetrahedron Lett. 1986,
2801. (d) Chenard, B.L.; Laganis, E.D.; Davidson, F.; RajanBabu, T.V. J. Org. Chem.
1985,50, 3666.
(74) (a) Mitchell, T.N.; Amamria, A.; Killing, H.; Rutschow, D. J. Organometallic
Chem., 1983,241, C45. (b) Mitchell, T.N.; Amarnria, A.; Killing, H.; Rutschow, D.
J. Organometallic Chem. 1986,304,257. (c) Mitchell, T.N.; Reimann, W. J.
Organometallic Chem., 1985,281, 163. (d) Mitchell, T.N.; Amamria, A. J.
Organometallic Chem., 1983, 252,47. (e) Killing, H.; Mitchell, T.N. Organometallics
1984,3, 1318.
(75) (a) Hayami, H.; Sato, M.; Kanemoto, S.; Morizawa, Y.; Oshima, K.; Nozaki, H.
J. Am. Chem. Soc. 1983,105,4491. (b) Wakamatsu, K.; Nonaka, T.; Okuda, Y .; Tiickmantel, W.; Oshima, K.; Utimoto, K.; Nozaki, H. Tetrahedron 1986,42,4427.
(c) Okuda, Y.; Morizawa, Y.; Oshima, K.; Nozaki, H. Tetrahedron Lett. 1984,2483.
(76) Okuda, Y.; Wakarnatsu, K.; ~ ~ c k m a n t e l , W.; Oshima, K.; Nozaki, H. Tetrahedron
Lett. 1985, 4629.
(77) (a) Watanabe, H.; Kobayashi, M.; Saito, M.; Nagai, Y. J . O~ganometallic Chem.
1980, 186. (b) Watanabe, H.; Kobayashi, M.; Saito, M.; Nagai, Y. J. Organometallic
Chem. 1981, 149.
(78) (a) Fugami, K.; Oshima, K.; Utimoto, K.; Nozaki, H. Tetrahedron Lett. 1986,
2161. (b) Fugami, K.; Nakatsukasa, S.; Oshima, K.; Utimoto, K.; Nozaki, H. Chem.
Lett. 1986, 869.
(79) (a) Koening, K.E.; Weber, W.P. Tetrahedron Lett. 1973,2533. (b) Miller, R.B.;
Reichenbach, T. Tetrahedron Lett. 1974, 543. (c) Brook, A.G.; Duff, J.M.; Reynolds,
W.F J. Organometallic Chem. 1976,121,293. (d) Jarvie, A.W.P.; Holt, A.;
Thompson, J. J. Chem. Soc. 1976, B, 852. (e) Fleming, I.; Pearce, J. J. Chem. Soc.,
Chem. Commun. 1975,633. (f') Calas, R.; Pillot, J.P. Bull. Soc. Chim. Fr. 1975,
2143. (g) Koening, K.E.; Weber, W.P. J. Am. Chem. Soc. 1973,95, 3416. (h) Chan,
T.H.; Lau, P.W.K.; Mychajlowskij, W. Tetrahedron Lett. 1977, 3317. (i) Fleming, I.;
Henning, R.; Plaut, H. J. Chem. Soc., Chem. Commun. 1984,29.
(80) (a) Stille, J.K. Angew. Chem. Int. Ed. Engl. 1986,25,508 and references cited
therein. (b) Mitchell, T.N.; Amamria, A. J. Organometallic Chem. 1981, 210, C17. (c)
Mitchell,.T.N.; Amamria, A. J. Organometallic Chem. 1983,47,252. (d) Seyferth, D.;
Weiner, M.A. J. Am. Chem. Soc. 1962,84, 361. (e) Seyferth, D.; Vaughan, L.G. J.
Am. Chem. Soc. 1964,86, 883. (f') Wulff, W.D.; Peterson, G.A.; Bauta, W.E.; Chan,
K.S.; Faron, K.L.; Gilbertson, S.R.; Kaesler, R.W.; Yang, D.C.; Murray, C.J. J.
Org. Chem. 1986,51, 277. Behling, J.R.; Babiak, K.A.; Ng, J.S.; Campbell, A.L.;
Moretti, R.; Koerner, M.; Lipshutz, B.H. J. Am. Chem. Soc. 1988,110, 2641.
(81) Kuivila, H.G. Synthesis 1970, 499.
(82) Gardette, M.; Alexakis, A.; Normant, J.F. Tetrahedron 1985,41,5887.
(83) (a) Campbell, J.B.; Brown, H.C. J. Org. Chem. 1980,45,550. (b) Molander,
G.A.; Brown, H.C. J. Org. Chem. 1981,46, 647. (c) Miyaura, N.; Itoh, M.; Suzuki,
A. Bull. Chem. Soc. Japan 1977,50,2199. (d) Campbell, J.B.; Brown, H.C. J. Org.
Chem. 1980,45, 549. (e) Yamamoto, Y.; Yatagai, H.; Maruyama, K.; Sonada, A.;
Murahashi, S-I. J. Am. Chem. Soc. 1977,99, 5652.
(84) (a) Zweifel, G.; Steele. R.B. J. Am. Chem. Soc. 1967,89, 2754. (b) Baba, S.;
Van Horn, D.E.; Negishi, E. Tetrahedron Lett. 1976, 1927. (c) Eisch, J.J.;
Damasevitz, J.E. J. Org. Chem. 1976,41,2214. (d) Uchida, K.; Utimoto, K.; Nozaki,
H. J. Org. Chem. 1976,41, 2215.
(85) Negishi, E.; Takahashi, T.; Akiyoshi, K. J. Chem. Soc., Chem. Commun. 1986,
1338.
(86) (a) Matsushita, H.; Negishi, E. J. Am. Chem. Soc. 1981,103, 2882. (b) Negishi,
E.; Takahashi, T.; Baba, S.; VanHorn, D.E.; Okukado, N. J. Am. Chem. Soc. 1987,
109, 2393. (c) Negishi, E.; Baba, S. J. Chem. Soc., Chem. Commun. 1976, 596. (d)
Baba, S.; Negishi, E. J. Am. Chem. Soc. 1976,98, 6729. (e) Negishi, E.; Valente,
L.F.; Kobayshi, M. J. Am. Chem. Soc. 1980,102, 3298. (f) Kobayashi, M.; Negishi,
E. J. Org. Chem. 1980,45, 5223.
(87) (a) Miyaura, N.; Ishiyama, T.; Ishikawa, M.; Suzuki, A. Tetrahedron Lett. 1986,
6369. (b) Miyaura, N.; Satoh, M.; Suzuki, A. Tetrahedron Lett. 1986,3745. (c)
Miyaura, N.; Satoh, M.; Suzuki, A. Tetrahedron Lett. 1980,2865.
(88) (a) Corey, E.J.; Beames, D.J. J. Am. Chem. Soc. 1972,94, 7210. (b) Jung,
M.E.; Light, L.A. Tetrahedron lett. 1982,3851.
(89) Piers, E.; Skerlj, R.T. J. Org. Chem. 1987,52, 4423.
(90) (a) Sheffy, F.K.; Godschalx, J.P.; S tille, J.K. J. Am. Chem. Soc. 1984,106,
4838. (b) Sheffy, F.K.; Stille, J.K. J. Am. Chem. Soc. 1983,105, 7173.
(91) Scott, W.J.; Crisp, G.T.; Stille, J.K. J. Am. Chem. Soc. 1984, 106, 4630.
(92) Stille, J.K.; Simpson, J.H. J. Am. Chem. Soc. 1987,109, 2138.
(93) Pierce, H.D.Jr.; Pierce, A.M.; Johnston, B.D.; Oehlschlager, A.C.; Borden, J.H.
J. Chem. Ecology 1988, 14,2169. (b) Johnston, B.D.; Oehlschlager, A.C. J. Org.
Chem. 1986,5I, 760.
(94) (a) Kobayashi, K; Kawanisi, M.; Hitomi, T.; Kozima, S. J. Organometallic Chem.
1982,233, 299. (b) Kitching, W.; Olszowy, H.A.; Drew, G.M. Organometallics
1982, I , 1244.
(95) (a) Hoffmann, D.M.; Hoffmann, R.; Fisel, C.R. J. Am. Chem. Soc. 1982,104,
3858. (b) Henerici-olive, G.; Olive, S. Coordination and Catalysis , Verlag Chirnie:
New York, 1977. (c) Bratsch, S.G. J. Chem. Ed. 1988,65, 34.
(96) Dewar, M.J.S. Bull. Soc. Chim. Fr. 1951,18, C79.
(97) Salomon, R.G.; Kochi, J.K. J. Organornet. Chem. 1974,64, 135.
(98) (a) Wrackrneyer, B. Annual Reports on NMR spectroscopy, Academic Press: New
York, Vol. 16,1985 and references cited therein. (b) Wrackrneyer, B. Polyhedron
1986,5, 1709.
(99) (a) House, H.O.; Watson, S.C.; Eastman, J.F. J. Organomet. Chem. 1967,9,
165. (b) Gilman, H.; Cartledge, F.K.; See-Yuen, S. J. Organomet. Chem. 1963,1, 8.
THE SYNTHESIS OF SULFONIUM MIMICS OF THE PRESUMPTIVE
CARBOCATION INTERMEDIATES IN THE DIMERIZATION OF
FARNESYL PYROPHOSPHATE TO SQUALENE BY SQUALENE
SYNTHETASE
Introduction
Excellent inhibition of enzymes may be achieved by administration of mimics of
presumptive intermediates that bind to catalytic sites but are not transformed. The
efficiency of intermediate mimics as inhibitors is due to the superior ability of enzymes to
stabilize intermediates formed during a catalyzed process.l Regulation of sterol
biosynthesis in man is a therapeutically valuable goal.2 A promising approach is the
inhibition of enzymes involved in sterol biosynthesis. Indeed the hypocholesteremic
agents most recently approved for human use are inhibitors of P-hydroxy-P-
methylglutaryl CoA reductase, a regulatory enzyme early in the sterol biosynthetic
process.3 As most other inhibitors of sterol biosynthesis they were discovered as natural
products and not by design.
The focus of the current work is the rational design of inhibitors of enzymes
involved in the biosynthesis of cholesterol that function after the formation of farnesyl
pyrophosphate but ate not involved in the biosynthesis of the ubiquinones (electron
transfer agents that are important in cell respiration). Unlike enzymatic carbon-carbon
bond formation leading to nucleic acids, proteins, carbohydrates and lipids that utilize
reactions proceeding through enolate equivalents, the enzymes involved in the advanced
stages of sterol biosynthesis utilize cationic processes for carbon-carbon bond
formation.4
Work in our and other laboratories during the last decade has made it apparent
that efficient inhibition of the latter enzymes can be achieved by administering sulfonium
and ammonium analogs of the presumptive cationic intermediates.596 Initial
investigations were aimed at design of ammonium and sulfonium intermediate analogues
of carbocations presumed to be involved in the alkylation of the side chain of zymosterol
(1) during biosynthesis of ergosterol (2) by the yeast Saccharomyces cerevisae.5
The enzyme responsible for this alkylation converts 1 to fucosterol3 (Scheme 1).
The mechanism suggested for yeast 24-sterol methyltransferase (24-SMT) involves
generation of cationic intermediates 4 and 5. Heteroatom analogues of these species
containing s u h 5 d (6,7 and 8) or nitrogensac (9 and 10) were excellent inhibitors
both in vim and in vivo (Scheme 2). The most efficient mimics were those containing a
sulfonium moiety at position 25 (compounds 6,7, and 8). Inhibitor 7 kii2 j;ciiZ 24-
SMT 25,000 times more tightly than the namal substrate for this enzyme and effectively
inhibited the enzyme at nanomolar concentrations (Ki -2 n~) .5d The superior ability of
224
7 compared to 8 as an inhibitor of 24-SMT suggested the stereochemistry of alkylation
was as illustrated in 4.5d
Scheme 1
It has also been shown by ~enveniste7 that 24-methyl 25-
azacycloartenol (l l) , 25- azacycloartenol(l2) and 25-methyl-25-azacycloartenol(13) as
well as arsonium and sulfonium analogs of 4 were potent inhibitors of cycloartenol24-
sterol methyltransferase in microsomes of maize seedlings (Scheme 3).
Scheme 2
Kinetic evidence suggests that the ionic heteroatom analogues of 4
exhibit their inhibitory power as reversible competitive inhibitors simply through
Coulombic attraction between the positively charged heteroatom and nucleophile(s)
nsponsible for stabilization of the presumptive carbocation intennediate.547
Scheme 3
This approach has led to the design of an ammonium inhibitor (14) for the
dimethyl ally1 pyrophosphate (15) to isopentenyl pyrophosphate (16) transformation8
(Scheme 4).
Nu: 7 H
1 4
Scheme 4
Similarly, sulfonium mimics (17 and 18) of presumptive cationic
intermediates (19 and 20) were efficient inhibitors of bomyl pyrophosphate cyclase and
a-pinene ~~clase .9 These enzymes accept geranyl pyrophosphate, 21, or linalyl
pyrophosphate, 22, as substrates and produce bomyl pyrophosphate, 23, and a-pinene,
24, respectively (Scheme 5). Inhibition by each sulfonium mimic is synergized (3-10 x)
by the addition of inorganic pyrophosphate confirming tight active site binding to the
postulated ion pair.
Squalene synthitase is a relatively small microsomal proteinlo that catalyses the
1'-1 condensation of two molecules of famesyl pyrophosphate (25) to yield squalene
(26). The transformations considered to be involved are the insertion of C1 of one
farnesyl pyrophosphate into the C2-C3 double bond of a second to generate intermediate
26 SQUALENE
Scheme 6
27 which rearranges to presqualene alcohol pyrophosphate, 28 (Scheme 6). Further
rearrangement of 28 via a cyclopropyl carbinyl rearrangement yields intermediate 29.
Ring opening of 29 yields allylic cation 30. Hydride reduction of the latter completes the
enzymatic transformation. Since the rearrangement of primary cyclopropylcarbinyl cation
28 to its tertiary isomer 29 was neither kinetically nor thermodynamically favoured in
solution it was assumed that squalene synthetase exerts strict control upon cationic
intermediates to achieve the regiocontrol required for biosynthesis of squalene. It was
suggested that regiocontrol was, in fact, achieved as a natural consequence of the
orientation between the positively and negatively charged partners in the tight ion pair
generated upon cleavage of the carbon-oxygen bond in 28. An ammonium analog, 31,
of intermediate, 29, has been shown to be an efficient inhibitor of squalene.ll
The preceeding examples reveal that ammonium and sulfonium analogues of
presumptive carbocationic intermediates are efficient inhibitors of enzymes presumed to
mediate processes via such intermediates. In all cases it is presumed that the positively
charged intermediates generated during the enzymatic reactions are stabilized by
nucleophilic sites located near the cationic carbons generated and that these nucleophilic
sites bind to the sulfonium and ammonium intermediate mimics through Coulombic
interactions.
Proposed work
The y s e n t study was undertaken to prepare the sulfonium ions 32,33 and 34
which are analogues of the cationic intermediates 27,29 and 30 respectively, presumed
to be stabilized by squalene synthetase (Scheme 7). It was anticipated that 32-34 would
be efficient inhibitors of this enzyme. Although there are several preparations of squalene
synthetase which could be studied, attention would be centered on the enzyme available
from yeast. The inhibition would be followed by incorporation of radiolabelled famesyl
pyrophosphate into squdene.12
R = /A Scheme 7
Results and discussion
Retrosynthetic analysis of 32 reveals that it can be prepared from fragments 35,
36 and 37. Fragment 35 is a homologated geraniol(38) derivative, whereas fragment
37 is a farnesol(39) derivative. The connecting hgment (36) can be visualized to be
derived from an a-haloester (Scheme 8).
32
Scheme 8
The thiol corresponding to 35 was prepared from geraniol(38) via
homogeraniol(40). The larter was prepared from 38 by the method developed by
~eo~old.l%"I'us, Swern oxidation14a of geraniol to g e m i d (41) produced less than
2% Z isomer. The aldehyde was converted to triene, 42, by Wittig reaction using methyl
triphenyl phosphoniurn iodide and phenyl lithium as the base.15 ~~droborationl6a of
42 with freshly prepand disamyl boranel6b gave 40 in 85% yield (Scheme 9).
Oxalyl chloride
DMSO, Et3N CH2C12, 91 % 4 1
Ph3PCH31 PhLi, THF
OH , HB(Siam),
NaOH, H202 4 0 85 % 4 2
Scheme 9
Alternate strategies to prepare homogeraniol(40) via geranyl cyanide
(43) were less satisfactory. The direct conversion of geraniol to geranyl cyanide failed in
our hands (Scheme 10).17 Although preparation of 43 from geranyl bromide (441~8
was
Bu3P, CCL4
38 NaI, NaCN 4 3
U B r COOH
Scheme 10
~uccessful,~9 it was abandoned because hydrolysis of the former gave appreciable
amounts of the Z isomer of 45 (Scheme 10).20
Homogeraniol(40) was converted to thiolacetate, 46, by the Mitsunobu
reaction.21 Reformation of the adduct between triphenyl phosphine and diisopropyl
azodicarboxylate is essential for the success of this reaction.22 Reduction of 46 by
lithium tetrahydridoaluminate gave the required thiol(47) in 53% overall yield (Scheme
11).23
47
Scheme 11
Famesyl bromide (48) was prepared from E3-famesol, 39, in an overall yield
of 54% (Scheme 12).18 The latter was prepared by the method of ~ e i l e r . ~ ~ At the time
this work was undertaken, E&-famesol was not commercially available.It is now
available from Aldrich in 98% isomeric purity. Ethyl acetoacetate was successively
treated with sodium hydride, n-butyl lithium and geranyl bromide, 44, to give 49 in
85% y i e ~ d . ~ ~ a Treatment of 49 with sodium hydride and diethyl phosphochloridate gave
enolphosphonate 5024b which was treated with mixed methyl cuprate, (formed by
treatment of cuprous iodide with methyl lithium and methyl magnesium bromide) to give
ethyl famesoate, 51.24~ Reduction of 51 with diisobutylaluminiwn hydride gave E,E - famesol, 39, in 93% yield.2%"I' was then converted to 48 in 90% yield by the
procedure of ~sbond. l8
I"'"
Scheme 12
Synthesis of sulfonium analogue 32 proceeded in 55% overall yield by coupling 47 and
48 (Scheme 13). Thiol47 was mated with thallium ethoxide26a and ethyl-a-homo
acetate according to the procedure of ~etty26b to produce 52 in nearly quantitative yield.
Treatment of 52 with lithium diisopropylamide27 followed by addition of 48 gave 53
in excellent yield. Use of the thallium salt of thiol47 (a soft base) avoided the 1,2-
addition reactions chaxtmeristic of the harder sodium and lithium thiolates. ~eduction23
of 53 with lithium tetrahydridoaluminate gave alcohol 54 in 98% yield. ~e th~la t iong of
54 with dimethyl sulfate gave 32 in 32% yield as a viscous oil (Scheme 13).
Scheme 13
Future directions
Direct synthesis of 53 can also be envisioned as shown in Scheme 14. Thus, the
dianion of thiomethyl ethyl acetate (55) can be generated by reaction with one equivalent
of NaH and one equivalent of n-butyl lithium.24a To this solution can be added one
equivalent of geranyl bromide, 44, followed by one equivalent of farnesyl bromide, 48.
It is visualized that reaction of geranyl bromide (44) and farnesyl bromide (48) with
dianion of 55 would be a facile process and should lead to the formation of 53 directly
from 44 and 48 in relatively fewer steps than the sequence in Scheme 13.
48
Scheme 14
SYNTHESIS OF CYCLOPROPYL SULFONIUM ION MIMIC, 33
Retrosynthetic analysis of sulfonium ion 33 revealed it could be fabricated from
the umpulong (57) of 35, a cyclopropyl carbinyl fragment (58) and a ylide (59) derived
from farnesol (Scheme 15).
33
Scheme 15
Umpulong of thiolate, 35, was envisioned as being derived from the
thiosulfonate ester, 60. All attempts to synthesize 60 from thiol47 and p-toluene
sulfonyl chloride failed (Scheme 16p8 and resulted in mostly disulfide, 61. Presumably
if 60 was formed from the anion of 47 and p-toluene sulfonyl chloride, it reacted with
remaining thiolate to produce 61. Inverse addition of the reagents did not significantly
alter the outcome of this process. Formation of the thiosulfonate ester via the tributyltin
intermediate 6228b was also unsuccessful in our hands. Thus, reaction of 62 ++".:ifi
p-toluene sulfonyl chloride yielded a complex mixture containing 61 along with a variety
of tin containing products (Scheme 16).
s-s L J
Scheme 16
Synthesis of 60 was achieved vin reaction of homogeraniol brornide,29 (63)
with sodium p-toluenesulfonyl thiolate, 64 as shown in Scheme 17.30
1 6 4 DMF, 84 %
Scheme 17
The ylide corresponding to synthon 59 (Scheme 15) was prepared from geranyl
bromide18 (44) by conversion to 65 (Scheme 18).31~ster 65 was converted to 66 by
reduction with lithium tetrahydridoaluminate. ~os~lation32a of 66 gave 67 which was
converted to the corresponding iodide (68) in good yield.32b Iodide 68 was mated with
mphenyl phosphine33 in benzene to form the Wittig salt, 69 in 38% yield. The low yield
was due to poor crystal formation. In a one pot sequence, 69 was methylated and
converted to the comsponding ylide, 70, according to the procedure of corry.34
i) LDA ii) CH3COOC2H5
0
Br -r
iii) CuI; -1 1 0 T o/\
44 88 % 65
p-TsC1 KOH
OTs - OH 67 79 % 6 6
Scheme 18
Synthesis of the cyclopropylcarbinyl fi-agment, 58 (Scheme 15), commenced
-OH MEMcl
Ir
Diisopropyl ethyl amine 72 %
I CHCl, NaOH CI'AB 89 %
Scheme 19
from the P methoxy-ethoxy-methoxy ether (71) of ally1 alcohol (Scheme 19).35 This
was converted to the dichlorocyclo propyl carbinyl intermediate, 72, by the procedure of
~oshi.36 Reaction of 72 and 60 proceeded smoothly to give 73 using the procedure of
de~oer.37 Dehalogenation of 73 to 74 was achieved by reaction with lithium tetra
hydroaluminate. Removal of the P-methoxyethoxy-methoxy ether of 74 to give 75 was
followed by Swem oxidation to give aldehyde 76 in 5 6 8 overall yield.14 Reaction of
76 and the ylide (70, vide supra) gave 77 as a 1: 1 mixture of cis,trans isomers with
respect to the two chains attached to the cyclopropane. Chromatographic separation of
isomers of 77 was not successful. Often unacceptably low yields (-5%) of the product
were obtained along with products whose NMR spectra indicated that the cyclapropane
ring had been destroyed, presumably because of the sensitivity of vinylcyclopropyl
moiety towards acidic Silica Gel.
Future directions
The formation of the cyclopropyl sulfonium ion (33) can be improved by the
addition of P ~ ~ S ~ C U to acetylene followed by protonolysis to produce high yields of
triphenylstannyl-1-ethene, 78, as shown in Scheme 20.38 Addition of the reagents
(CHC13, NaOH, CTAB, vide rnpra)36 to the alkene, 78, would result in the
corresponding dichlom cyclopropyl intermediate 79. Introduction of fragment 57
followed by &halogenation as described above should result in the formation of 80.
Transmetallation of triphenylstannyl moiety followed by conversion of the resultant
cyclopropyl-lithium reagent into the comspondhg organozinc chloride (81) would be
achieved using the procedure of ~krs.39 In situ F~itTirjFj4 ~camiymi wupiing of 81
with the vinyl iodide 82 should give 77 of the desired stereochemistry.39
Ph3SnCu = H ~ H CHC13, NaOH =
M a f l H SnPh, CTAB - SnPh,
78 73
1.2 equiv. n-BuLi, 1
-78OC. 30 min
1.2 equiv. ZnCl2, -40•‹C, 30 min
0.02 equiv. Pd(Ph-J4 'Y
77 Scheme 20
The vinyl iodide, 82, can be prepared by the dianion of acetylene with coupling
of 63 in the presence of CuCN to yield the alkynyl precursor, 83.40 Carboalurnination-
iodination of 83 should result in the stereospecific formation of 82 (Scheme 21).41 This
sequence not only utilizes novel reactions but would also avoid the use of
chromatographic separation in the preparation of 77 since the desired stereochemistry is
fixed in 80.
Scheme 21
STEREOSPECIFIC SYNTHESIS OF VINYL SULFONIUM MIMIC, 34
The several methods to stereospecifically prepare vinyl sulfides may be divided
into two principle strategies. One involves the creation of an alkynyl sulfide followed by
stereospecific reduction. Procedures available to prepare alkynyl sulfides involve the
reaction of akynyl anions with sources of electrophilic sulfur (Scheme 22a)$2 direct
displacement of akynyl halogen by nucleophilic s u b (Scheme 22b).42 Reduction of
alkynyl sulfides with aluminum hydride =agents is reported to yield E or Z vinyl sulfides
depending upon the reagent used (Scheme 22~).43
90 % R'-C-Li+ + CH3SCN - R'-S-CH3 (a)
CH3S0 CH3 CH,SS~H,
Scheme 22
The second strategy involves reaction of a vinyl organometallic
with electrophilic sulfur. Several metals have been successfully applied in this strategy.
Thus, diisobutyl-aluminum hydride undergoes stereospecific cis hydroalumination with
allqnes to give E-vinyl alanes (Scheme 23).43a These may be converted to vinyl sulfides
by a number of routes:
a) Reaction of E-vinyl alanes with dkyl lithium *agents yield intermediate
alanates which react with electrophilic sulf~r.4~b
b) Reaction of E-vinyl alanes with bromine produce vinyl bromides which may
be reacted with nucleophilic sulfur directly in the presence of HMPA (Scheme
2 3 b ) k or viu PdO catalysis (Scheme U C ) . ~
Scheme 23
Alternatively, vinyl bromides may be converted to vinyl Grignards
(Scheme 23d) and thence to vinyl sulfoxides via sulfinic acid esters. Vinyl sulfoxides
are easily reduced to vinyl sulfides via reaction with ethyl Grignard and cuprous
iodide.44e
Carbocupration of alkynes also provides a stereospecific route to E disubstituted
and trisubstituted vinyl cuprates (Scheme 24). These react with both disulfides and
thiosulfonate esters to yield vinyl sulfides with retention of stere~chemistry.~~
Scheme 24
Recently, Corey has &vised a stereospecific route to divinyl
sulfides involving stereospecific generation of a vinyl thiolate anion via reaction of vinyl
lithium intermediates with styrene sulfide. The vinyl thiolates generated coupled
stereospecifically with vinyl iodides when reacted as the cuprous thiolates (Scheme
25).45
"wH BuLi
R SnBu3 R Li .
"wH BuLi 4
R S'LP R
, CuI U
Scheme 25
Retrosynthetic analysis of 34 reveals that this sulfonium mimic of 30 can be
prepared by combination of synthon 57 and the vinyl anion equivalent 84 (Scheme 26).
Scheme 26
Reaction of farnesyl bromide (48) and dilithioacetylide in the presence of
CuCN gave 85, the required alkynyl precursor of synthon 84 (Scheme 271.m Coupling
the anion of 85 with thiosulfonate ester 60 yielded the alkynyl sulfide 86 which upon
reduction with lithium tetrahydridoaluminate gave the required E-vinyl sulfide, 87
(Scheme 27). Methylation of 87 with CH3I gave the sulfonium ion mimic 34 in 27%
yield.
We attempted to react the thiosulfonate ester with vinyl organometallic in order to
generate vinyl sulfide 87 directly from the alkynyl precursor 85 (Scheme 28). Reaction
of diisobutyl aluminum hydride with 85 gave mixtures of partial and fully reduced
products and was abandoned.448 Reaction of diisoamyl borane with 85 cleanly gave the
vinyl borane product as judged by hydrolytic workup of the initial reaction mixture.16
Conversion of the presumed E-1-vinyl borane (88) to vinyl sulfide 87 was attempted
under conditions reported by suzuki46 for cross coupling reactions of vinyl boranes.
One reaction used NaOH to generate a boronate anion which, on reaction with
thiosulfonate ester, 60, was expected to yield 87. This reaction consistently gave
disulfide (61) and
Scheme 27
thioether (89) as the niain products, presumably from competing =action of base with
thioester. Addition of (Ph3P)qPd under similar reaction conditions gave the desired vinyl
sulfide, albeit in low yield (-15%).
NaOH B(Siamh
88
I
Scheme 28
250
Future directions
The results to date suggest that vinyl sulfides are formed via transition metal
catalyzed reaction of thiosulfonates with vinyl boranes. This is consistent with literature
on PdO catalyzed coupling of vinyl boranes with alkenyl halides in the presence of base.
Experiments suggested are:
a) Reaction of vinyl borrmes with PdO catalyst in the presence of nonhydroxylic
base such as NaOMe or tertiary amines. Since catechol-boranes or 9-BBN
function better in PdO catalyzed crosscoupling reactions than diisoamyl
boranes, reactions should be attempted with vinyl boranes derived from the
above reagents.
b) To ensure PdO remains in the proper oxidation state, a reducing species such
as CuBn should be added to the reaction.
c) In order to achieve oxidative addition of Pd to electrophilic sulfur to obtain
RS-Pd-X, perhaps, RSX or RSSR should be ~sed as a sulfur source. Use of
a weaker base in PdO cataly,zed cross coupling eaction of vinyl borons with
thiosulfonate esters would provide a definite synthetic advantage over the
existing routes.
EXPERIMENTAL
1~ NMR spectra were recorded on a Bruker WM-400 spectrometer in CDCl3.
Mass spectra were obtained on a Hewlett-Packard 5985B GUMS system at 70eV.
Infrared spectra were recorded neat on a Perkin Elmer Model 599B spectrophotometer.
Elemental analyses were performed by Mr. M. Yang of Simon Fraser University,
Department of Biological Sciences on a Perkin Elmer Model 240 elemental analyzer. Gas
chromatographic analysis were performed on a Hewlett-Packard 5880A instrument using
a J and W fused silica DB-1 column (15 m x 0.25 mm id.), equipped with a flame
ionization detector. Thin layer chromatography was conducted on aluminium sheets
precoated with 60 F254 Silica Gel (E Merck, Darmstadt). All column chromatography
was performed on Silica Gel 60 (230-400 mesh, E Merck, Darmstadt) as described by
still.47
Tetrahydrofuran was ksh ly distilled from potassium benzophenone-ketyl,
dimethyl sulfoxide, hexamethyl phosphorous triamide, diisopmpyl amine, triethyl mine
and dimethyl f d d e were distilled from CaH2 and stored over molecular sieves (3
A). Dichloromethane was distilled from P205 and stored over molecular sieves (4 A).
Unless otherwise stated all reactions were conducted under argon in flame dried
glassware. The general work up procedure was as follows: the reaction mixture was
quenched with an ice cold 10% solution of W C l , the aqueous layer was extracted with
ether (3 x 50 rnL), the combined organic phase was washed with saturated NaCl
solution, dried over anhyd. MgSO4 filtered and concentrated in vacua
Preparation of (E)-3,7-Dimethyl-2,6-octadienal (41). To a stirred
solution of oxalyl chloride (29.2 g, 0.23 mol) in CH2C12 (500 mL) was rapidly added
dimethyl sulfoxide (34.3 g, 0.44 mol) in CH2Cl2 (100 mL) at -7S•‹C. After stining for 5
min at the same temperature, geraniol(38,30.8 g, 0.2 mol) was added dropwise over
10 min. After stirring for 15 min, triethylamine (140 mL, 1.0 mol) was added dropwise
while maintaining the temperature below -60•‹C. The mixture was stirred for 5 min, then
warmed to rt before addition of 100 mL of water. The aqueous layer was extracted with
(2 x 300 mL) portions of CH2Cl2. The organic layer was washed with 1% HCl until no
longer basic, then washed successively with 50 mL portions of water, 5% Na2C03,
water and brine. The combined organic extract was dried over anhyd MgS04 and
concentrated in vacuo to give 34.7 g of the crude product. Flash chromatographic
(hexanedethyl acetate, 411) purification gave 27.6 g (91%) of geranial41. Gas
chromatographic analysis revealed a purity of 98% with less than 2% Z isomer. 1~
NMR (CDCl3) 6 1.56 (s, 3H, vinyl methyl), 1.72 (s, 3H, vinyl methyl), 2.10 (s, j ~ ,
C3-vinyl methyl), 2.18 (m, 4H, CH2), 5.15 (t, lHy Cg-H), 5.82 (d, lH, C2-H, J = 10
Hz), 9.2 (d, 1Hy C1-H, J = 10 Hz). MS, m/e (%), 153.2 (0.3), 152.3 (3.2), 94.2
(14.1), 84.2 (25.5), 41.1 (41.6) and 69.1 (100). IR , 2920, 2720, 1680, 1390, 990 and
820.
Preparation of (E)-4,8-Dimet hyl-1,3,7-nonatriene (42). To methyl
triphenylphosphonium iodide (38.9 g, .096 mol) in 250 mL of THF was added of 2.4 M
phenyl lithium dropwise at O•‹C until a permanent yellow color appeared (1 mL). Then
phenyl lithium (37.3 mL, 0.089 mol) was added while keeping the temperature below
O•‹C. This procedure ensures that the reaction mixture is anhydrous. The deep red
solution of ylide containing excess phosphonium salt was stirred for 0.5 h at rt. Geranial
(41, 13.4 g, 0.088 mol) in THF (40 mL) was added dropwise at O•‹C. The reaction was
warmed to rt and stirred overnight after which a light orange suspension was formed.
The reaction was quenched by adding 2 mL of methanol. Most of the solvent was
evaporated in vacuo. and the resulting slurry was diluted with pentane (200 mL) and
decanted onto 50 g of Celite. The solids were washed with hot pentane (3 x 50 mL) and
filtered through Celite. The resulting yellow solution was filtered through Florosil(50 g)
to remove colored impurities. The organic solution was dried over anhyd. MgS04 and
concentrated in vacuo to yield 19.96 g of crude product. Distillation yielded 12.1 g
(91%) of 42, b.p. 23-24OC 0.05 mm Hg (Lit. b.p. 22-24OC 0.05 mm Hg). Gas
chromatographic analysis revealed a purity of 98% with presence of 1.6% Z isomer. 1~
NMR (CDCl3) 6 1.60 (s, 3H, vinyl methyl), 1.68 (s, 3H, vinyl methyl), 1.76 (s, 3H,
C4-vinyl methyl), 2.03-2.14 (my 4H, CH2), 4.96-5.11 (my 3H, C1, 9-vinyl H), 5.85
(d, lH, C3-vinyl H, J = 10.5 Hz), 6.65 (td, lH, C2-vinyl H, J = 10.5 Hz, 16.89 Hz).
MS, m/e (%), 151.2 (0.4), 150.2 (3.4), 135.2 (4.1), 81.2 (14.5), 79.2 (15.2), 41.1
(26.1) and 69.1 (100.0). IR , 3035, 1650, 1600, 1450, 1380,990, 900 and 810.
Preparation of (E)-4,8-Dimethy1-3,7-nonadien-l-o1 (40). To diborane
(50 mL, 50 mmol) in THF at -30•‹C was added rapidly 2-methyl-2-butene (15.4 g, 0.22
mol). The reaction was stirred at O•‹C for 2 h. Concurrently, a solution of 42 (6.75 g, 45
mmol) in THF (20 mL) was added to another flask under argon. The disamyl borane
was transferred via a canula to the addition funnel of the second flask. The temperature
of both reactions was held at O•‹C during the 1 hr addition. The reaction was stirred at O•‹C
for an additional hr, then overnight at n. Excess disamyl borane was destroyed by
addition of methanol (1 mL). After cooling to O•‹C, 3 M NaOH (17 mL) was added
rapidly. Then 17 mL of 30% H202 was added while the temperature was maintained
below - 10•‹C. The reaction was stirred at rt for 3 h. Usual workup gave 1 1.1 1 g of crude
product Distillation yielded 6.37 g (85%) of homogeraniol, 40, b.p. 65-67OC O.lmm
Hg. Gas chromatographic analysis revealed a single component. 1~ NMR (CDC13) 6
1.57 (s, 3H, vinyl methyl), 1.61 (s, 3H, vinyl methyl), 1.65 (s, 3H, Cq-vinyl methyl),
1.96-2.07 (m, 4H, CH2), 2.22-2.27 (q, 2H, C2-CH2 J = 6.6 HZ), 3.56 (t, 2H, C 1 - CH2 J = 6.6 HZ), 5.03-5.11 (m, 2H, C3, C7-vinyl H). MS, m/e (%), 168.3 (0. I), 125
(33.5)- 81.1 (22.7), 69.0 (loo), 41.1 (21.5). IR , 3320, 2920, 1665, 1440, 1375,
11 10, 1050,875,830 and 740.
Syrrthesis of 1-Bromo-(E)-3,7-dimethyl-2,6-octadiene (44).
Phosphorus tribromide (3.42 g, 1 1.9 mmol) was added dropwise to an ice cooled
solution of geraniol(38,4.45 g, 28.8 mmol) in 25 mL of anhyd. ether. After 3 hr the
reaction was quenched with ice-cold saturated NaHC03. Usual workup gave 5.8 g
(93%) of 44. Thin layer chromatography using hexane/EtOAc (411) as the eluant
revealed a s.ingle spot. IH NMR (CDCl3) 6 1.57 (s, 3H, vinyl methyl), 1.67 (s, 3H,
vinyl methyl), 2.0-2.13 (m, 4H, CH2), 4.2 (d, 2H, C1-CH2 J = 7 Hz), 5.04-5.1 (m,
1H, C6-vinyl H), 5.2 (t, lHy C2-vinyl H J = 7 Hz).
Preparation of (E)-4,8-Dimethyl-3,7-nonadienonitrile (43). Method
A. To an efficiently stirred solution of KCN (0.65 1 g, 10 mmol) and 18-crown-6 ether
(0.132 g, 0.5 mmol) in HMPA (15 mL) was added a solution of geraniol(0.77 g, 5.0
mmol) and tri-n- butyl phosphine (1.11 g, 5.5 mmol) in 5 mL of CH3CN over 10 min.
The flask was cooled in an icelmethanol bath and CCl4 (0.85 g, 5.5 mmol) was added.
The reaction was stirred for 3 days after which time it was diluted with Et20 (200 mL)
and washed with 1Wo aq. citric acid (50 mL). The usual workup and purification by
flash chromatography on silica using (4/1), hexane/EtOAc (Rf = 0.23), gave 0.8 13 g
(67%) of nitrile 43. Gas chromatographic analysis revealed a purity of 97%. IN'MR
(CDC13) 6 1.56 (s, 3H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.72 (s, 3H, vinyl
methyl), 2.02-2.19 (my 4H, CH2), 3.5 (d, 2H, Cg-CH2 J = 7.4 Hz), 5.04-5.2 (m, 2H,
vinyl H). MS, xde (%), 164.3 (0.1), 163.3 (0.9), 162.2 (0.8), 149.2 (0.5), 123.2
(6.5), 67.2 (9.0), 41.1 (25.0) and 69.2 (100). IR , 2926, 2731, 2248, 1701, 1655,
1574, 1515,1231, 1082,984 and 827.
Method B. To an efficiently stirred solution of geraniol(1.0 g, 6.5 mmol),
NaCN (0.645 g, 13.0 mmol) and NaI (2-5 mg) in CH3CN (10 mL) was added a
solution of Me3SiCl(1.40 g, 15.0 mmol) in DMF (10 mL). The reaction mixture was
refluxed far 8 hr after which time it was poured onto water (100 mL). The usual workup
and purification by flash chromatography on silica using (4/1), hexane/EtOAc (Rf =
0.23), gave 0.255 g (21%) of nitrile 43. The spectral analysis was as reported above.
Method C. To a stirred solution of geranyl bromide, 44, (5.35 g, 24.6 mmol,
prepared as described below) in CH3CN (20 mL) was sucessively added 18-crown-6-
ether (0.25 g, 0.95 mmol) and KCN (4 g, 61.4 mmol). The reaction was stirred in the
dark for 6 days after which time it was filtered. The solvent was removed and the residue
was titurated with hexane:EtOAc (3: I), to separate the 18-wn-6 ether. The solids were
removed by filtration and the filtrate concentrated in vacua Distillation of the crude
product gave 2.4 g (59.7%) of 43, b.p. 90-91•‹C 0.2- Hg. Gas chromatographic
analysis revealed a single component. The spectral analysis was identical to that given
above.
Synthesis of (E)-4,s-Dimethyl-3,7-nonadienoic acid (45). Nitrile, 4 3
(3.5 g, 21.5 mmol) was dissolved in MeOH (30 mL), and an aqueous KOH solution
[4.0 g (71 mmol) in 8 mL of H20] was added. The reaction mixture was refluxed for 48
h, cooled, diluted with NaHCQ, extracted with EQO, acidifkd with 2N HCl, filtered
and concentrated to give crude homogeranic acid (45).as a light brown oil (3.2 g, 82%).
IH NMR (CDCl3) 6 1.57 (s, 3H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 2.0-2.13
(m, 4H, CH2), 3.06 (d, 2H, C1-CH2 J = 7 Hz), 5.04-5.1 (m, lH, Cg-vinyl H), 5.2 (t,
lH, C2-vinyl H J = 7 Hz). IR ,2400-3600, 1725.
Preparation of (E)-4,8-Dimethyl-3,7-nonadien-1-thioacetate (46).
To an efficiently stirred solution of triphenylphosphine (21 g, 80 mmol) in THF (200
mL) was added diisopropylazodicarboxylate (16.7 g, 80 mmol) at O•‹C. The reaction was
stirred for 0.5 hr, after which time a thick white precipitate was obtained. A mixture of
thiolacetic acid (6.1 g, 80 mmol) and 40 (6.72 g, 40 mmol) in 100 mL THF was added
dropwise while the temperature was maintained below O•‹C. The reaction was stirred for
1.5 hr at O•‹C and then overnight at rt. Solvent was removed in vacw from the clear
orange solution and the resulting product purified by flash chromatography using
hexane:CH2Cl2 (3: I), or hexane:EtOAc (49: I), as eluant. Distillation gave 7.54 g (83%)
of 46, b.p. 65-67OC 0.Olmm Hg. Gas chromatographic analysis showed a single
component. IH NMR (CDC13) 6 1.59 (s, 3H, vinyl methyl), 1.60 (s, 3H, vinyl
methyl), 1.67 (s, 3H, Cq-vinyl methyl), 1.95-2.06 (m, 4H, CH2), 2.22-2.27 (q, 2H,
C2-CH2 J = 7.3 Hz), 2.31 (s, 3H, COCH3), 2.84-2.87 (t, 2H, C1-CH2 J = 7.3 Hz),
5.05-5.12 (m, 2H, C3, C7-vinyl H). MS, m/e (%), 226.2 (0.1), 186.2 (0.1), 185.1
(0.6), 184.2 (IS), 81.1 (59.3, 69.0 (loo), 43.1 (42.5), 41.1 (45.3). IR , 3350, 2910,
1690,1435,1350,1130,1100,950,830 and 735. Anal. calcd for C13H22S0, C 69.02
H 9.73. Found C 69.22 H9.75.
Preparation of (E)-4,8-Dimet hyl-3,7-nonadien-1-thiol (47). To a
stirred solution of lithium tetrahydridoaluininate (1.22 g, 8.0 equiv) in anhyd. ether (30
mL) was added dropwise a solution of 46 (5 . ~ d g, 32. I -01) in 13 mL of anhyd.
ether. The reaction was stirred for 0.5 hr, then cooled to O•‹C. Excess hydride was
destroyed by careN addition of ice-cold 1N HCl. Precipitated salts were removed by
filtration through 2 cm of Celite. Usual workup gave 7.5 g of crude product. Distillation
gave 5.42 g (92%) of 47, b.p. 46OC 0.Olmm Hg. Gas chromatographic analysis
revealed a purity of 96%. 1~ NMR (CDC13) 6 1.40-1.44 (t, 1H. SH J = 7.6 Hz), 1.59
(s, 3H, vinyl methyl), 1.61 (s, 3H, vinyl methyl), 1.67 (s, 3H, C4-vinyl methyl), 1.98-
2.02 (my 2H, C6-CH2), 2.04-2.08 (my 2H, C5-CH2), 2.28-2.33 (q, 2H, C2-CH2 J =
7.1 Hz), 2.49-2,54 (dt, 2H, C1-CH2 J=7.1, 7.5 Hz), 5.07-5.11 (my 2H, C3, C7-
vinyl H). Irradiation of the triplet at 6 1.40- 1.44 resulted in the simplification of the
signal at 2.49-2.54 (C1-CHz), thus c o n f i i g the signal at 1.40-1.44 to be that of the
SH. MS, m/e (%), 186.2 (0.1), 185.2 (0.2), 141.1 (96.9), 81.1 (36.5), 69.2 (100) and
41.1 (41.0). Isobutylene, CI, m/e (%), 186.0 (12.48)- 185.0 (100), 151.0 (20.8), 95.0
(16.64). IR ,2920,2560 (w), 1735, 1660, 1440, 1285, 1110,980 and 830.
Synthesis of Ethyl [(E)-7,11-dimethyl-3-keto-6,10-dodecadien]-1-
oate (49). Ethyl acetoacetate (2.7 g, 21 mmol) was M e d dropwise to a stirred solution
of NaH (1.31 g, 23.5 mmol, 57% mineral oil) in 50 mL of THF at -lO•‹C. The reaction
was stirred for 20 min when n-BuLi (10 mL, 21 mmol,) was added slowly while the
temperature was maintained below -5OC. The resulting yellow solution was stirred for 20
min before the addition of 44 (5 g, 23.1 mmol) in 5 mL of THF. The reaction was
stirred for 1 hr before it was quenched with ice-cold saturated NHqCl. The usual work
up gave 5.8 g of crude product which was distilled, b.p. 112-1 14OC O.1mm Hg, to yield
4.6 g (84%) of 49. Gas chromatographic analysis revealed a purity of 93%. IH NMR
(CDC13) 61.30 (t, 3H, OCH2CH3 J = 10 Hz), 1.62 (s, 6H, vinyl methyl), 1.70 (s, 3H,
Cq-vinyl methyl), 1.95-2.10 (my Cg, C6-CH2), 2.30 (q, 2H, C2-CH2 J = 6.6 Hz),
2.58 (t, 2H, C1 -CH2 J = 6.7 HZ), 3.45 (s, 2H, CO-CH2-CO), 4.20 (q, 2H,
OCH2CH3 J = 10 Hz), 5.06-5.12 (m, 2H, C3, C7-vinyl H). MS, m/e (%), 266 (1 8.8),
248 (25), 223 (56.2), 205 (18.8), 136 (25), 109 (87.5), 81 (31.2), 69 (loo), 41 (62.5).
IR, 2920, 1735,1630, 1450, 1310, 1235, 1035,910 and 840.
Synthesis of Ethyl [(E)-7,ll-dimethyl-3-(diethoxy phosphonate)-
2,6,10-dodecatrienl-1-oate (50). To NaH (0.3 13 g, 57% mineral oil) in 10 mL of
anhyd. ether was added P-keto ester 49 (1.33 g, 5 mmol) in 5 mL ether at O•‹C. The
reaction was stirred for 20 min before diethylphosphochloridate (0.97 g, 5.6 mmol) was
added dropwise. The resulting mixture was stirred for 2 hr and quenched with saturated
W C l . The usual work up procedure resulted in crude phosphonate ester which was
purified by flash chromatography using hexaneEtOAc (311). The yield of 50 was 1.76 g
(88%). Gas chromatographic analysis revealed a single component. IH NMR (CDCl3) 6
1.25 (t, 3H, COOCH2CH3 J = 7.0 Hz), 1.35 (t, 6H, P(OCH2CH3)2 J = 7.0 Hz), 1.60
(s, 6H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.92-2.10 (m, 4H, C7, Cg-CH2),
2.24 (q, 2H, C4-CH2 J = 7.1 Hz), 2.46 (t, 2H, C3-CH2 J = 7.12 Hz), 4.15 (q, 2H,
COOCH2CH3 J = 7.0 Hz), 4.25 (quint, 4H, P(OCH2CH3)2 J = 7.0 Hz), 5.05-5.15
(m, 2H, vinyl H), 5.36 (s, lH, C1-vinyl H). MS, m/e (%), 402 (12.5), 357 (31.25).
287 (25), 220 (31.25), 155 (loo), 99 (50), 69 (43.75) and 41 (37.5). IR ,2940,2870,
1732,1670,1450, 1290,1210, 1040 and 825.
Preparation of Ethyl [(E)-3,7,11-trimethyl-2,6,10-dodecatrienl-1-
oate (51). To CuI (1.42 g, 7.46 mmol) in 15 mL THF at -lO•‹C was added dropwise
with stirring MeLi (4.7 mL, 7.46 mmol), while the temperature was maintained below - 10•‹C. Afler 30 min MeMgBr (4.0 mL, 12.43 mmol) was added dropwise at -40•‹C. The
reaction was further stirred for 0.5 h before addition of 50 (1 g, 2.48 mmol) in 5 mL of
THF. The reaction was stirred for 1.5 hr before it was warmed to rt. The reaction was
quenched by addition or" i~iiiE:i;i~K&Xi (9;; j. The usual workup and evaporation of
solvent in vacuo gave 0.78 g of crude product, which was purified by flash
chromatography using hexane/EtOAc, 5011) as eluant (Rf = 0.29). Gas chromatographic
analysis revealed a purity of 94%. The yield of 51 was 0.57 g (87%). IH NMR
(CDC13) 6 1.25- 1.28 (t, 3H, OCH2CH3 J = 7.1 Hz), 1.59 (s, 6H, vinyl methyl), 1.67
(s, 3H, vinyl methyl), 1.95-2.06 (my 8H, Cq, Cg, C8, Cg-CHz), 2.15 (s, 3H, C3-vinyl
methyl), 4.1 1-4.16 (q, 2H, OCH2CH3 J = 7.1 Hz), 5.07-5.09 (my 2H, vinyl H), 5.66
(s, 1Hy C2-vinyl H). MSy d e (%), 264 (12.5), 221 (37.5),128 (56.25), 81 (62.5) and
69 (100). IR ,2924,1725,1651,1450, 1230 and 1151.
Synthesis of (E&)-3,7,11-Trimethyl-2,6,l0-dodecatrien-l-ol (39).
DIBALH (40 mL, 40 mmol) was added dropwise to an efficiently stirred solution of 51
(3.55 g, 13.44 mmol) in 25 mL of anhyd. ether at -50•‹C. The reaction was stirred at the
same temperature for 2 hr after which time it was quenched by addition of 300 mL of
10% aqueous tartaric acid. The usual workup, followed by evaporation of the solvent in
vocuo gave 2.76 g (93%) of E,E-famesol(39) in 96% purity. IH NMR 6 1.59 (s, 6H.
vinyl methyl), 1.67 (s, 6H, vinyl methyl), 1.95-2.12 (my 8H, C4, Cg, C8, Cg-CH2),
4.13-4.16 (t, 2H, Cl-CH2 J = 6.15 Hz), 5.06-5.12 (my 2H, C6, C10-vinyl H), 5.40-
5.43 ( t lH, c 2 - h y l H J = 6.2 Hz). MS, Isobutylene CI, m/e (%), 221 (12.5), 205
(100), 149 (41.6) and 137 (83.3). IR ,3340 (b), 2940, 1450, 1010 and 850.
Preparation of 1-Bromo-(E&)-3,7,11-trimethyl-2,6,10-
dodecatriene (48). To a stirred solution of E,E-farnesol (39,6.21 g, 0.028 mol) in
25 mL of ether was added dropwise phosphorus hibromide (3.24 g, 0.012 mol) at - 10•‹C. The reaction was stirred for 3 h after which time it was quenched by pouring into
an ice-cold saturated solution of NaHC03. The usual workup, followed by removal of
solvent in vucuo , yielded 7.16 g (90%) of 48. TLC using hexane/EtOAc (411) revealed
a single component ( ~ f = u.85). :H ~n4.R (CDC13) S 1.59 (s, 6H, vinyl methyl), 1.67
(s, 3H, vinyl methyl), 1.75 (s, 3H, C3-vinyl methyl), 1.95-2.19 (m, 8H, Cq, Cg, Cg,
Cg-CHz), 4.0-4.15 (d, 2H, Cl-CH2 J = 7.1 Hz), 5.05-5.12 (my 2H, Cg, C10-vinyl
H), 5.5-5.57 (t, lH, C2-vinyl H J = 7.1 Hz).
Synthesis of Ethyl [(E)-4,s-dimethyl-3,7-nonadienl-1-thioacetate
(52). To a Schlenk tube, connected via a two-way stopcock to an argon inlet and a
vacuum line was added 47 (4.05 g, 22 mmol) in 20 mL of etherhexme (1/1). To this
stirred solution was added thallium ethoxide (5 g, 20 mmol) via a syringe. Immediate
yellow coloration was observed. The reaction was stirred far 5 min before the solvent
was removed under vacuum. The resulting yellow oily residue was washed with ether (2
x 10 mL), the solvent removed under vacuum and the residue redissolved in 50 mL of
anhyd. ether. Ethyl-a-homo acetate (3.32 g, 20 mmol) was added dropwise whereupon
a white precipitate of thallous bromide was formed. After stirring for 1 hr at rt the
reaction was filtered through 2 cm of Celite. After the removal of solvent, the product
was purified by flash chromatography using hexane/dichlmmethane (211) to give 52 as
a single component (Rfd.325) as the eluant, 5.3 g (97%) of 96% purity as revealed by
gas chromatographic analysis. IH NMR (CDCI3) 6 1.28 (t, 3H, OCH2CH3 J = 6.7
HZ), 1.59 (s, 3H, vinyl methyl), 1.66 (s, 3H, vinyl methyl), 1.66 (s, 3H, C4-vinyl
methyl), 1.96-2.1 (m, 4H, CHZJ), 2.3 (q, 2H, C2-CH2 J = 7.1 Hz), 2.61 (t, 2H, C 1-
CH2 J = 7.1 Hz), 3.25 (s, 2H, -SCH2CO), 4.2 (q, 2H, OCH2CH3 J = 6.6 Hz), 5.06-
5.16 (m, 2H, C3, C7-vinyl H). MS, m/e (%), 272 (2.0), 270.6 (10.0), 201 (30), 183
(9.0), 150 (22), 107 (29), 81 (58.5), 69 (100) and 41 (41.5). IR 2980,2960, 1735,
1450,1380,1275,1130, 1030 and 835. Anal. Calcd for C15H2602S: C, 66.66, H,
9.63. Found: C, 66.46, H, 9.90.
Synthesis of Ethyl-[2-(4',8'-dimethyl-3'(E),7'-nonadienyl-thio)-
5,9,:3-ti~~i~;Z~:-5(Z;,~,:2-tetradecatrien]oate (53). To diisopropylamine (0.5
mL, 3.5 mmol) in 5 mL THF was added dropwise at O•‹C, n-BuLi (1.7 mL, 3.5 mmol)
in hexane. The reaction was stirred for 15 min before addition of 52 (0.94 g, 3.5 mmol)
in 5 mL of THF. The resulting mixture was s t . for 2.5 hr at O•‹C then cooled to -50•‹C
and stirred for an additional 0.5 hr. Farnesyl bromide (48, 1.4 g, 5 mmol) was then
added. Tht reaction was stirred for 1 hr at -50•‹C and warmed to rt whereupon it was
quenched by pouring onto-ice-cold saturated NH&l solution. The usual work up and
purification by flash chromatography using hexaneICH2Cl2 (Yl) gave 1.52 g (92%) of
53 (Rf = 0.34) as eluant. IH NMR (CDC13) 6 1.27 (t, 3H, -COOCH2CH3 J = 7.1 Hz),
1.59 (s, 15H, vinyl methyl), 1.67 (s, 6H, vinyl methyl), 1.92-2.08 (m, 12H, C5', Cg',
C6, C7, C10, C ~ I - C H ~ ) , 2.25-2.62 (m, 6H, Cl', C2, C3-CH2), 3.23-3.27 (m, lH,
SCHCOO), 4.16-4.19 (m, 2H, COOCHZCH~), 5.06-5.13 (m, 5H, C3', C7', Cq, Cg,
C12 vinyl H). Irradiation of signal at 6 4.16-4.19 resulted in a singlet at 6 1.27 (triplet
origudy) and simplified the signal at 6 3.23-3.27. This showed the vicinal coupling
between CH2 and CH3 and long range coupling of CHI with SCHCOO. MS, m/e (%),
475 (0.2), 474 (0.6), 405 (4.8), 183 (3.0), 95 (14.8), 81 (25.9), 69 (100.0) and 41
(57.0). IR ,2930,1735,1450, 1380, 1 150,1110 and 1040. Anal. Calcd for
C30H5002S: C, 75.90, H, 10.55. Found: C, 75.64, H, 10.74.
Synthesis of 2-[(4',8'-dimethyl-3'(E),7'-nonadienyl-thio)-(5,9,13-
trimethyl-4(E),8(E),12-tetradecatrien)]-l-ol (54). To an efficiently stirred
solution of LiAZHq (0.039 g, 1.04 mmol) in 7 mL of ether was added thioacetate 53
(0.49 g, 1.04 mmol). The reaction was stirred for 1 hr before it was quenched by careful
addition of ice-cold 1N HCl. The precipitated salts were removed by filtration through 2
cm of Celite and the filtrate subjected to normal work up procedure to give 4.33 g (98%)
of 54. Thin layer chromatography using hexane/EtOAc (4/1) as eluant revealed a single
component. IH NMR (CDCl3) 6 1.59 (s, 12H, vinyl methyl), 1.61 (s, 3H, vinyl
methyl), 1.67 (s, 6H, vinyl methyl), 1.95-2.1 (m, 12H, C5', C6', C6, C7, C10, C11-
CH2), 2.22-2.31 (m, 4H, C2', C3-CH2), 2.48-2.55 (m, 2H, C1-CH2), 2.75-2.79 (m,
lH, SCHCH20H J = 12.5, 6.80, 4.62 HZ), 3.44-3.50 (ddd, lH, HCHaOH J = 11.8,
6.2, 5.8 HZ), 3.63-3.69 (ddd, 1H, HCHbOH J = 11.4, 7.1, 4.54 Hz), 5.06-5.14 (my
5H, vinyl H). MS, m/e (%), 432 (6.5), 363 (80.54), 213 (6.5), 183 (4.4), 81 (15.2),
69 (100.0), and 41 (17.4). IR ,3440,2920, 2860, 1450, 1380, 1030 and 835. Anal.
Calcd for C28H480S C, 77.78, H, 11.11. Found C, 77.53, H, 11.36.
Synthesis of 32. The thiol (54, 0.21g, 0.5 mmol) was dissolved in acid-free
dimethyl sulfate (5 mL, distilled under vacuum) in a sealed test tube for 2 h at 90•‹C.
Following complete consumption of the thiol, as judged by TLC on Silica Gel
(hexane/EtOAc, 411)- the resulting dark solution was washed several times with anhyd.
Et2O and the insoluble residue dissolved in H20 (10 mL). The material was precipitated
by addition of perchloric acid and recrystallized from water containing a few percent of
perchloric acid to give (0.07 g, 32%) of the sulfonium salt, 32. 1~ NMR (CD30D) 6
1.64 (brs, 15H, vinyl methyl), 1.67 (brs, 6H, vinyl methyl), 2.3-2.5 (my 16H, C2',
C3, C5', C6', C6, C7, Cloy C11-CHz), 2.7 (my 2H, Cl-CH2), 2.8 (my 1Hy
SCHCH20H), 2.95 (s, CH3), 3.57 (ddd, lH, HCHaOH J = 11.8, 6.2, 5.8 HZ), 3.74
(ddd, 1Hy HCHbOH J = 11.4, 7.1, 4.54 Hz), 5.06-5.14 (my 5H, vinyl H). MS, FAB,
(Xenon/sulfolane) m/e 447 (M+, 100).
Attempted syntheses of 60. Synthesis of thiosulfonate ester 60 using p-
toluene sulfonyl chloride in combination with NaH, Py or Et3N and thiol47 was
attempted. The major product was &sulfide 61. A general procedure follows:
Normal addition: Thiol47 (0.368 g, 2 mmol) was added dropwise to a
suspension of NaH (0.14 g, 2.5 mmol, 57% oil) in 5 mL of THF. After 30 min of
stirring, p-toluene sulfonyl chloride (0.38 g, 2 mmol) in 10 mL of THF, was added
dropwise. The reaction was stirred for 3 hr and then subjected to normal work up.
Purification of the product by flash chromatography using h e x a n e / a 1 2 (311) as
eluant yielded 0.51 g (76%) of disulfide 61 and 0.05 g (7.4%) of 60.
Inverse addition: The reaction was conducted in the same manner as above
except that the sodium salt of 47 was added to the p-toluene sulfonyl chloride. No
appreciable change in yield was observed (10%).
Spectral data for 61: IH NMR (CDC13) 6 1.59 (s, 6H, vinyl methyl), 1.61 (s,
6H, vinyl methyl), 1.67 (s, 6H, vinyl methyl), 1.95-2.10 (my 8H, Qj, Qj', Cg, C5'-
CH2), 2.35 (q, 4H, C2, C2'-CH2 J = 7.0 HZ), 2.55-2.6 (t, 2H, C1, C1'-CH2 J = 7.1
Hz), 5.04-5.1 (my 4H, Vinyl H). MS, Isobutylene CI, d e (%), 367 (30.0), 339 (30.0),
241 (25.0), 225 (32.5), 207 (52.5), 185 (22.5), 183 (100.0), 151 (25.0).
Preparation of Tri-n-butyl-(E,E)-4,s-dimethyl-3,7-nonadien-1-
thio)stannane (62). To an efficiently stirred solution of thiol47 (6.07 g, 33 mmol)
and triethylamine (4 g, 39.0 mmol) in 200 mL of CC14, was added dropwise n-Bu3SnC1
(10.73 g, 33.0 mmol). The reaction w& stirred overnight at n, filtered and the filtrate
washed with 5% aqueous acetic acid, then water (50 mL). The organic layer was
separated and dried over anhyd Na2S04. Flash evaporation of the solvent gave needle
shaped white crystals, which were recrystallized from hexane. 1~ NMR (CDCl3) 6 0.86
(t, 9H, CH3, J = 7.1 Hz), 1.2- 1.4 (my 12H, CH2), 1.4- 1.5 (m, 6H, CH2), 1.59 (s,
6H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.95-2.10 (my 4H, C6, Cg, -CH2), 2.35
(q, 2H, C2, -CH2 J = 7.0 Hz), 2.55-2.6 (t, 2H, C1, -CH2 J = 7.1 Hz), 5.04-5.1 (my
2H, Vinyl H). MS, d e (%), 416 (M+ - 57,22).
Preparation of Sodium thio-p-toluenesulfonate (64). To Na2Se9H20
(24 g, 0.1 mol) in 20 mL of water cooled to O•‹C was added to p-toluene sulfonyl
chloride (19.05 g, 0.1 mol) while the temperature was maintained below 5OC. After
completion of addition, the reaction was warmed to 85OC until all deposited sulfur
dissolved. Solvent was removed under reduced pressure and the resulting powder
washed with 50 mL of ethg then extracted with hot absolute EtOH. The product was
recrystallized using absolute EtOH and the crystals dried over P205 at 120•‹C for 12 hr.
This procedure yielded 24 g (98%) of 64. 1~ NMR (D2O) 6 2.39 (s, 3H, phenyl-
CH3), 7.35-7.78 (AA'BB', 4H, phenyl H). IR , 1330, 1200, 1090,980, and 800(b).
A similar pmedure was used to prepare NaSSO2CH3. IH NMR @20) 6 2.77
(s).
Synthesis of l-Bromo-4,8-dimethy1-3(E),7-nonadiene (63). Bromine
was added dropwise to an efficiently stirred ice-cooled solution of triphenylphosphine
(3.2 g, 12.2 mmol) in 20 mL of CH2Cl2 until a permanent yellow color appeared. A few
milligrams of triphenylphophine were added to consume excess Br;! (i.e., until no
yellow coloration remained). At this point 2.0 mL of pyridine was added and the reaction
stirred for 10 min. Homogeraniol(40,2.0 g, 12 mmol) in 20 mL of CH2C12 was added
dropwise over 30 min. The reaction was warmed to rt and stirred for 1 hr then filtered.
The precipitate was washed with pentane, filtered, and further washed with water (2 x 50
mL). The solvent was dried over anhyd. MgS04, concentrated in vauo and the residue
subjected to column chromatography using hexane as eluant. The yield of 63 (Rf =
0.47) was 2.1 g (78%). Gas chromatographic analysis revealed a purity of 99%. 1~
NMR (CDCl3) 6 1.57 (s, 3H, vinyl methyl), 1.61 (s, 3H, vinyl methyl), 1.65 (s, 3H,
vinyl methyl), 1.96-2.13 (m, 4H, Cg, C6-CH2), 2.53-2.62 (q, 2H, C2-CH2 J = 6.67
Hz), 3.32-3.38 (t, 2H, C1-CH2 J = 6.66 Hz), 5.03-5.1 1 (m, 2H, vinyl H). MS, m/e
Synthesis of (4,7-Dimethyl-3(E),7-nonadien-1-thio)-p-
toluenesul fona te (60). To an efficiently stined solution of NaS S02C6H5CH3 (64,
2.45 g, 10 mmol) in 30 mL of DMF was added homogeranyl bromide (63,2.31 g, 10
mmol) in 10 mL of DMF. The reaction was stirred at rt for 3 days after which time it was
quenched by pouring into 100 mL of water. The usual workup and purification by
column chromatography using hexane/CH2Cl2 (1/1) as eluant gave 2.83 g (84%) of
thiotosylate 60. Gas chromatographic analysis revealed a purity of 9 1 % and the presence
of 3% of disufide 61. IH NMR (CDC13) 6 1.55 (s, 3H, vinyl methyl), 1.61 (s, 3H,
vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.92-2.06 (my 4H, Cg, C6-CH2), 2.28-2.35
(q, 2H, C2-CH2 J = 6.7 Hz), 2.46 (s, 3H, phenyl-CHs), 2.97-3.02 (t, 2H, Cl-CH2 J
= 6.7 HZ), 4.9-5.08 (m, 2H, vinyl H), 7.32-7.84 (AA'BB', 4H, phenyl H). MS, m/e
(%), 339 (6.3), 255 (25.2), 183 (100.0), 117 (43.7), 115 (38), 69 (87.5). IR ,2920,
1670, 1600, 1500, 1450(b), 1330(b), 1150, 1080, and 830.
Preparation of Ethyl 5,9-dimethyl-4(E)&decadienoate (65). To
diisopropylamine (4.65 g, 46 mmol) in 30 mL THF at -65OC was added dropwise n-
BuLi (17.7 mL, 46 mmol). The reaction was stirred for 1.5 hr after which time it was
transferred via syringe to another flask containing a mixture of ethyl acetate (4.1 g, 46
mmol) and CuI (17.5 g, 92 mmol) in 175 mL of THF at -1 10•‹C. The reaction was
warmed to -30•‹C, whereupon geranyl bromide 44 (5 g, 23 mmol) in 60 mL of THF was
added slowly to maintain the temperature at -30•‹C. The reaction was stirred for 1 hr at the
same temperature after which time it was subjected to the normal workup procedure.
Distillation yielded 4.52 g (88%) of 65. Gas chromatographic analysis showed a purity
of 91%. IH NMR (CDCl3) 6 1.22- 1.25 (t, 3H, COOCH2CH3 J = 7.1 Hz), 1.58 (s,
3H, vinyl methyl), 1.60 (s, 3H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.94-2.05
(my 4H, CH2), 2.3 (s, 4H, C1, C2-CH2), 4.08-4.13 (q, 2H, COOCH2CH3 J = 7.1
Hz), 5.04-5.10 (my 2H, vinyl H). MS, d e (%), 224.2 (1.9), 219 (0.5), 210.3 (1.8),
209.3 (10.2), 94.1 (84.6), 100.1 (92.7), 81.1 (97.4) and 95.2 (100.0). IR , 2980,
2920, 1740, 1450, 1370, 1150, 1050 and 835.
Preparation of 5,9-Dimethyl-4(E),8-decatrien-1-01 (66). To an
efficiently stirred solution of (0.68 g, 17.84 mmol) in 20 mL of ether was added
dropwise 65 (4.04 g, 18 mmol) in 30 mL of ether. After 30 min excess hydride was
destroyed by careful addition of 1N HC1. The solution was filtered through 2 cm of
Celite. Usual workup procedure gave 3.12 g (95%) of 65 in 98% purity as measured by
gas chromatographic analysis. IH NMR (CDCl3) 6 1.58 (s, 3H. vinyl methyl), 1.59 (s,
3H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.95-2.09 (my 8H, C2, C3, Qj, C7-
CH2), 3.61 -3.64 (t, 2 8 C1 -CH2 J = 6.6 Hz), 5.05-5.13 (m, 2H, vinyl H). MS, m/e
(%), 183.2 (0.1), 182.2 (0.7), 139.2 (100.0), 95.1 (84.1), 69.1 (76.2) and 67.2
(7 1 .5).
Preparation of 5,9-Dimethyl-4(E),8-decadien-1-p-toluene sulfonate
(67). Powdered KOH (13 g, 232 mmol) was added in 5 g portions to a solution of 66
(3.07 g, 16.86 mmol) andp-toluene sulfonyl chloride (3.69 g, 19.25 mmol) in 40 mL of
ether at -30•‹C. The resulting suspension was stirred for 2 hr at O•‹C after which time it
was quenched by pouring onto ice water (50 mL). The usual workup yielded 4.45 g
(79%) of 67 as found by gas chromatographic analysis. This material was used in the
next reaction without further purification. IH NMR (CDCl3) 6 1.57 (s, 3H, vinyl
methyl), 1.67 (s, 3H, vinyl methyl), 1.87-2.08 (m, 8Hy C2, C3, C6, C7-CH2), 2.43
(s, 3H, phenyl-CH3), 3.99 (t, 2H, C1-CH2 J = 7.1 Hz), 4.92-5.07 (my 2H, vinyl H),
7.3-7.8 (AABB', 4H, phenyl H). MS, m/e (%), 337.3 (0.2), 336.3 (0.8), 322.6 (0.2),
321.3 (0.7)- 121.1 (92.7)- 95.1 (100.0) and 69.1 (84.1). IR , 2920, 1450, 1350(b),
1180,96O(w) and 835.
Synthesis of 1-Iodo-5,9-dimethyL4(E),&decadiene (68). A solution
of NaI (4.32 g, 28.8 mmol) in 40 mL of acetone was brought to reflux under argon with
stirring. To this solution-was added 67 (4.38 g, 13 mmol). After 1 hr at reflux the
reaction was worked up in the usual fashion using pentane for extraction. The product
was purified by column chromatography using hexane as eluant. The yield of 68 (Rf =
0.45) was 3.72 g (98%). Gas chromatographic analysis revealed a purity of 90%. 1~
NMR (CDC13) 8 1.57 (s, 3H, vinyl methyl), 1.59 (s, 3H, vinyl methyl), 1.67 (s, 3H,
vinyl methyl), 1.82-1.86 (q, 2H, C2-CH2 J = 7 Hz),1.95-2.12 (my 6H, C3, C6, C7 - CH2), 3.15-3.20 (t, 2H, C1-CH2 J = 7.0 Hz), 5.03-5.10 (my 2H, vinyl H). MS, m/e
(%), 293.1 (0.2), 292.2 (1.4)- 279.2 (0. l), 249.0 (69.2), 95.0 (88.8), 69.0 (100.0)
and 41.1 (65.1).
Synthesis of (~j-5,9-~imeth~l-4,8-decadien~l)-l-
triphenylphosphonium iodide (69). A solution of 68 (1.55 g, 5.3 mmol) and
triphenylphosphine (1.7 g, 6.5 mmol) in 2 mL of benzene was placed in the dark for 5
days. The resulting solution was added to 100 inL of ether with rapid stirring. Fine white
crystals were obtained which were filtered and washed with ether (3 x 10 mL).
Recrystallization using benzenekther (1/1) gave 1.12 g (38%) of 69. IH NMR (C(j&j)
8 1.59 (s, 3H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.76 (s, 3H, vinyl methyl),
2.09-2.21 (my 6H, C3, C6, C7-CH2), 2.70-2.78 (q, 2H, C2-CH2 J = 7.0 Hz), 4.55-
4.65 (my 2H, C1-CHz), 5.17-5.25 (my 2H, Cq, Cg-vinyl H), 7.19 (my 15H, phenyl
H). MS, FABy (xenon/sulfolane) (%), 427 (M+,100.0), 289 (373,275 (25.0), 262
(72.9).
Preparation of 71. To a freshly distilled solution of pethoxy-methoxy methyl
chloride (4.98 g, 0.039 mmol) in 150 mL of CH2C12 was added diisopropyl axnine (5.1
g, 0.039 mmol) at O•‹C. The reaction was stirred for 10 min after which ally1 alcohol
(2.09 g, 0.036 mmol) was added over 0.5 hr. The reaction was stirred for 18 hr and
worked up in usual fashion. Distillation gave 3.78 g (72%) of 71, b.p. 60-61•‹C 17 mm
Hg. Gas chromatographic analysis revealed a single component. 1~ NMR (CDCl3) 6
3.4 (s, 3H, OCH3), 3.55-3.75 (AA'BB', 4H, OCH2CH20), 4.10 (td, 2H, C3-CH2 J
= 6.5 Hz, 3.0 Hz), 4.76 (s, 2H, OCH20), 5.17-5.32 (my 2H, vinyl H), 5.88-5.98 (ddt,
1Hy C2-vinyl H J = 18 Hz, 10 Hz, 6.5 Hz). MS, Isobutylene CI, m/e (%), 147 (14.6),
89 (100.0). IR ,2910,2890,2150, 1450,910 and 745.
Preparation of 72. To a solution of 71 (9.34 g, 0.064 mol) in 25 mL of
CHC13 was added acetyl trimethylammonium bromide (0.2 g, 0.55 mmol) and a chilled
50% NaOH solution (125 g of NaOH and 125 mL of water). The two phase suspension
was stirred vigorously and analyzed periodically by gas chromatography (linear
temperature program from 40'-200•‹C at 10•‹/min) over 3 days. After this time the reaction
ceased and mixture was diluted with 100 m . of water and the CHC13 layer separated.
The aqueous layer was extracted with CH2C12 (3 x 20 mL) and the combined organic
extracts washed with water to which MgS04 was added to break an emulsion during
washing. The organic layer was dried over anhyd.MgSO4, the solvent removed in vacuo
and the residue distilled to yield 12.3 g (89%) of 72, b.p. 58O-59OC 0.025 mm Hg. Gas
chromatographic analysis revealed a single component. IH NMR (CDC13) 6 1.20- 1 .25
(t, lHy C ~ C ~ O P I Q P ~ ~ H J = 8.89 Hz), 1.63-1.67 (dt, lHy cyclop~~pyl H J = 8.9 Hz, 2.9
Hz), 1.92-2.0 (dddd, 1 H, cyclopropyl H J = 5.6 Hz, 4.5 Hz, 2.7 Hz and 2.2 Hz); 3.37
(s, 3H, OCH3), 3.55-3.60 (AA'BB, 2H, OCH2CH2-CH2), 3.6-3.65 (dd, lH, carbinyl
Ha J = 9.8 Hz, 8.4 Hz), 3.73-3.77 (AA'BB, 3H, OCH2CH20 and carbinyl Hb), 4.74-
4.76 (s, 2H, OCH20). MS, Isobutylene CI, m/e (%), 231 (10.41), 229 (18.75), 125
(373, 123 (56.25), 105 (100). IR ,2890, 1455, 1400, 1155, 1055 and 755. Anal.
calcd for C8H1403C12 C, 41.92, H, 6.61. Found C, 41.89, Hy 6.16.
Preparation of 73. Dichlorocyclopropyl derivative 72 (0.5 g, 2.31 mmol)
was added to 15 mL of ether and the reaction cooled to -lOO•‹C (liquid N2JEtOH sluny).
n-BuLi (1.25 mL, 3 mmol) was added dropwise to maintain the temperature below - 100•‹C. After stirring for 2 h at -100OC thiotosylate 60 (1.23 g, 3.65 mmol) was added
dropwise and the temperature was kept below - 10O0C. The reaction was stirred for 30
min and then warmed to rt whereupon it was diluted with 5 mL of water and extracted
with pentane (3 x 10 mL). The extract was dried over anhydNa2SOq and concentrated
in vacuo. The residue was purified by column chromatography using hexaneEt0Ac
(611) (Rf = 0.32) as eluant. Two fractions were collected.
Fraction 1 (0.08gY 0.34 mmol) , suggested.= BuSCllHlg, was codlrmed by
IH NMR (CDC13) 6 0.94 (t, 3H, Bu-CH3 J = 7.2 Hz), 1.4 (sixtet, 2H, Bu-CH2 J =
7.2 HZ), 1.57 (my 4H, Bu-CHz), 1.57 (s, 3H, vinyl methyl), 1.59 (s, 3H, vinyl
methyl), 1.67 (s, 3H, vinyl methyl), 1.97-2.1 (m, 4H, Cg, Q-CHz), 2.28 (q, 2H, C2-
CH2 J = 7.1 Hz), 2.54 (q, 2H, C1 -CH2 J = 7.1 Hz), 5.07-5.2 (my 2H, C3, C7-CH2).
Fraction 2 (0.7gY 1.87 mmol) and the major one was 73. IH NMR (CDCl3) 6
1.35 (t, lH, cyclopropyl H J = 9.0 Hz), 1.60 (my lH, cyclopropyl H), 1.62 (s, 3H,
vinyl methyl), 1.98-2.2 (my 5H, Cg, Q-CH2 and cyclopropyl H), 2.38 (q, 2H, C2-
CH2 J = 7 HZ), 2.86 (t, 2H, C1-CH2 J = 7.1 Hz), 3.4 (s, 3H, OCH3), 3.55-3.6
(AA'BB', 2H, OCH2CH20), 3.62-3.65 (dd,lH, carbinyl Ha , J 9.8 Hz, 8.5 Hz),
3.73-3.8 (m, 3H, OCH2CH20 and c w i y l Hb), 4.78 (s, 2H, OCH20), 5.07-5.10 (m,
2H, vinyl H).
Since cyclopropyl proton signals were obscured by other signals, the 1~ NMR
spectrum was obtained in C@6. 1~ NMR (C-) 6 0.74 (t,lH, cyclopropyl H J = 6
HZ), 1.35 (dd, IH, cyclopropyl H J = 6 Hz, 8 Hz), 1.59 (s, 3H, vinyl methyl), 1.61 (s,
3H, vinyl methyl), 1.74 (s, 3H, vinyl methyl), 1.99-2.24 (m, 5H, Cg, C6-CH2 and
cyclopropyl H), 2.4 (q, 2H, C2-CH2 J = 7 Hz), 2.98 ( dt, 2H, Cl-CH2 J = 7 Hz, 2
HZ), 3.15 (s, 3H, OCHs), 3.40 (t, 2H, OCH2CH20 J = 5 Hz), 3.56 (dd, lH, carbinyl
Ha, J = 9.4 Hz, 8 Hz), 3.62-3.74 (m, 3H, OCHlCH20 and carbinyl Hb), 4.65 (s, 2H,
OCH20), 5.3 (m, 2H, vinyl H).
Confixmation of the assignment of the signal at 6 1.98 to be to Ha and those at 6
0.74 and 6 1.35 to be due to Hb was achieved by spin decoupling experiments.
Irradiation at 6 0.74 reduced the dd at 6 1.35 to a doublet (J = 8 Hz). This also simplified
the multiplet at 6 1.98-2.20. Irradiation at 6 1.98 reduced the triplet at 6 0.74 to a doublet
(J = 6 Hz) and the dd to a doublet (J = 6 Hz). The dd at 6 3.56 was also reduced to a
doublet (J = 9.4 Hz) and,the signal 6 3.62-3.74 was simplified. . .
MS, m/e (%) 377 (20.8), 341 (25), 301 (97.9), 265 (45.8), 15 1 (100). Anal.
calcd for for ClgH3303SCl C, 60.55, H, 8.76. Found C, 60.61, H, 8.69.
Synthesis of 74. To of LiAlH4 (0.15gY 4 mmo1)in 5 mL of dry Et20,73
(1.13g, 3.0 mmol) was added and the mixture refluxed for 3 h. It was then quenched
carefully with water and extracted with pentane. Usual workup followed by column
chromatographic purification (hexane/EtOAc, 6/1) gave 75% (0.77g) of the desired
p r o d ~ t . 'H NMR (C@6) 6 0.54 (d~lHy ~ y c ; u p u ~ y i ii i' = 6 EL, XZ), 0.74 (td,lH,
cyclopropyl H J = 6 Hz, 8Hz), 1.35 (dt, lH, cyclopropyl H J = 6 Hz, 8 Hz), 1.59 (s,
3H, vinyl methyl), 1.61 (s, 3H, vinyl methyl), 1.74 (s, 3H, vinyl methyl), 1.99-2.24
(m, 5H, Cg, C6-CH2 and cyclopropyl H), 2.4 (q, 2H, C2-CH2 J = 7 Hz), 2.98 ( dt,
2% Cl-CH2 J = 7 Hz, 2 Hz), 3.15 (s, 3H, OCH3), 3.40 (t, 2H, OCH2CH20 J = 5
Hz), 3.56 (dd, lH, carbinyl Ha, J = 9.4 Hz, 8 Hz), 3.62-3.74 (m, 3H, OCHlCH20
and carbinyl Hb), 4.65 (s, 2H, OCH20), 5.3 (m, 2H, vinyl H). MS, xn/e (%) 342
(43.8), 303 (loo), 287 (40.3),253 (76.3), 69 (100).
Synthesis of 77. The synthesis of this cyclopropyl analog (77) commenced
with the preparation of the cyclopropyl alcohol derivative (75) and subsequent oxidation
using the procedure developed by Swem et a1.14 Wittig olefination of the aldehyde with
the phosphonium ylide (70, generated in situ) from the phosphonium iodide (70)
completed the sequence. Thus, 74, (0.34g, 1.0 mmol) was added to 6 mL of a mixture
of THEH20:AcOH (1: 1:3) at rt. After stirring for 4 h, the reaction was extracted with
pentane. Usual workup followed by removal of solvent yielded 75 which was
immediately oxidized to the aldehyde 76. To a stirred solution of oxalyl chloride (0.292
g, 2.3 mmd) in CH2Cl2 (5 mL) was rapidly added dimethyl sulfoxide (0.343 g, 4.4
mrnol) in CH2Cl2 (1 mL) at -78OC. After stirring for 5 min at the same temperature, 75,
0.508 g, 2.0 mmol) was added dropwise over 10 min. After stirring for 15 min,
triethylamine (1.4 mL, 0.1 mmol) was added dropwise while maintaining the
temperam below -60•‹C. The mixture was stirred for 5 min, then warmed to rt before
addition of 1 mL of water. The aqueous layer was extracted with (2 x 3 mL) portions of
CH2C12. The organic layer was washed with 1% HCl until no longer basic, then washed
successively with 5 mL portions of water, 5% Na2C03, water and brine. The combined
organic extract was dried over anhyd. MgS04 and concentrated in vacuo to give 0.44 g
(87%) of the crude product. IH NMR (CDC13) 6 1.48-1.52 (my lH, cyclopropyl H),
1.57 (s, 6H, vinyl methyl), 1.57-1.62 (my lH, cyclopropyl H), 1.67 (s, 3H, vinyl
methyl), 1.98-2.1 (my 4H, Cg, C&CH2), 2.21-2.26(m, lH, cyclopropyl H), 2.38 (q,
2H, C2-CH2 J = 7 Hz), 2.41 (my lH, cyclopropyl H), 2.86 (t, 2H, C1-CH2 J = 7.1
Hz), 5.07-5.10 (my 2H, vinyl H) 9.25 (d, 1Hy CHO, J 3 Hz). MSy d e (%) 254 (20.2),
151.0 (20.8), 95.0 (16.64). IR , 2920, 1725, 1440, 1320, 1285, 1110,980 and 830.
Anal. Calcd for C15H260S C, 70.86, H, 10.20. Found C, 70.53, H, 10.16.
A solution of 0.54g (0.97 mmol) of 69 in 3 rnL of anhydrous THF was cooled
to O•‹C. n-BuLi (0.97 mmol, 0.4 mL) was added, and the resulting solution was allowed
to warm to rt Stirring was continued for 10 min, methyl iodide (0. l4g, 0.97 mmol) was
added dropwise, and the resulting solution was stirred for 30 min at rt. The reaction
mixture was cooled to O•‹C, and a second portion of n-BuLi (0.97 mrnol, 0.42 mL) was
added. The solution was allowed to warm to rt and then stirred for 10 min. The reaction
mixture was cooled to -45OC, and 0.248 g (0.98 mmol) of 76 was added via a syrnge.
Stirring was continued for 15 min at this temperature. Et20 (1 mL) was added and the
solution allowed to warm to O•‹C. After 0.5 h of stirring, absolute EtOH (1 mL) was
added. The resulting clear yellow solution was allowed to warm to rt and was stirred
overnight. The solution was then poured onto 10 mL of Et20 and extracted with water
until the aqueous layer was pH 7. Usual workup followed by flash chromatography on
Silica Gel (benzenehexme, 1/1) resulted in four fractions. Those fractions containing the
pure trans isomer (77) were pooled to yield O.08g (0.196 mmol, 20% from 76) Rf =
0.56; IH NMR (CDC13) 6 0.7- 1.1 (m, lH, cyclopropyl H), 1.59 (s, 12H, vinyl
methyl), 1.57- 1.60 (my 1 H, cyclopropyl H), 1.6 1 (s, 3H, vinyl methyl), 1.67 (s, 6H,
vinyl methyl), 1.95-2.1 (my 10H,-CHz), 2.22-2.31 (m, 4H, C2, C3-CH2), 2.48-2.55
(my 2H, Cl-CH2), 2.21-2.26 (my lH, cyclopropyl H), 2.38 (q, 2H, C3-CH2 J = 7
Hz), 2.4 1 (my 1 H, cyclopropyl H), 4.64 (d, lH, vinyl H, J 9 Hz, lHz), 5.07-5.10 (m,
4H, vinyl H). MSy m/e (%) 435 (20.2).
Those fractions containing the pure cis isomer (77) were pooled to yield O.08g
(0.196 mmol, 20% from 76) Rf = 0.61; 1~ NMR (CDC13) 6 0.7- 1.1 (my lH,
cyclopropyl H), 1.59 (s, 12H, vinyl methyl), 1.57-1.60 (m, 1Hy cyclopropyl H), 1.64
(s, 9H, vinyl methyl), 1.95-2.1 (m, 10H,-CH2), 2.22-2.31 (m, 4H, C2, C3-CH2),
2.48-2.55 (my 2H, Cl-CH2), 2.21-2.26 (my lH, cyclopropyl H), 2.38 (q, 2H, C3-
CH2 J = 7'Hz), 2.41 (m, lH, cyclopropyl H), 4.54 (d, lH, vinyl H, J 9 Hz, lHz),
5.07-5.10 (m, 4H, vinyl H). MS, m/e (%) 435 (34.2).
Synthesis of 5,9,13-trimethyl-4(E),8,12-tetradecatrien-l-yne (85).
A stream of dry acetylene (passed successively over CaCl2, and two dry ice acetone
traps) was used to saturate 50 mL of anhyd.THF at -50•‹C. The temperature was lowered
to -78OC and n-BuLi (3.05 mL, 6.4 mmol) was added dropwise while the temperature
was maintained below -70•‹C. The reaction was stirred for 30 min then CuCN (0.286 g,
3.2 mmol) in 10 mL of THF was added in one portion. The tan suspension was further
stirred for 30 rnin at -78OC and then at O•‹C for 1 hr when it was cooled again to -70•‹C.
Farnesyl bromide 48 (0.71 g, 2.5 mmol) in 5 mL of anhyd.THF, was then added
dropwise. The reaction was stirred at this temperature for 1 hr after which time it was
warmed to 0•‹C and further stirred for 1 hr. The usual work up and column
chromatography using hexane as eluant (Rf =0.37) gave 0.55 g (96.49%) of 85. Gas
chromatographic analysis showed a single component IH NMR (CDC13) 6 1.59 (s,
6H, vinyl methyl), 1.62 (s, 3H, vinyl methyl), 1.67 (s, 3H, vinyl methyl), 1.95-1.97 (t,
lH, acetylene H J = 2.7 Hz), 1.98-2.10 (m, 8Hy C6, C7, C10, C11- CH2), 2.88-2.9 1
(ddt, 2H, C3-CH2 J = 6.8 Hz, 2.7 Hz, 0.808 Hz), 5.08-5.15 (m, 2H, Cg, C12-vinyl
H), 5.17-5.2 (dt, lH, Cq-vinyl H J = 0.8 Hz, 2.7 Hz), MS, m/e (%), 230 (3.3), 215
(20), 187 (26.6)- 136 (26.6), 105 (20), 9 1 (33.3), 8 1 (46.6), 69 (100). Anal. Calcd for
C17H26 C, 88.69, H, 11.30. Found C, 88.65, H, 11.31.
Preparation of 86. To alkyne 85 (0.143 g, 0.625 mmol) in 5 mL of THF
under argon and cooled to -5OC was added with stirring n-BuLi 0.26 mL, 0.625 rnmol)
dropwise while the temperature was maintained at -5OC. The reddish orange solution
obtained was stirred for 30 min then thiotosylate 60 (0.17 g, 0.5 mmol) was added
dropwise. The reaction was stirred for 1 hr after which time it was subjected to narmal
work up and column chromatogrphy using hexane/CH2Cl2 (1511) as eluant. The yield of
86 (Rf =0.29) was 0.17 g (82%). IH NMR (CDCl3) 8 1.60 (s, 6H, vinyl methyl), 1.62
(s, 9Hy vinyl methyl), 1.70 (s, 6H, vinyl methyl), 1.97-2.15 (my 12H, C5', Cgl, C6,
C7, C10, C ~ I - C H ~ ) , 2.45 (dt, 2H, CpCH2 J = 7.1 Hz, 7.0 Hz), 2.67 (t, 2H, C1-
CH2 J = 7.1 Hz), 5.08-5.2 (my 5H, vinyl H). MS, Isobutylene, CI, m/e (%) 413
(42.8), 335 (14.3), 317 (9.9, 275 (23.8), 221 (19.0), 207 (14.3), 185 (23.8), 151
(100). MS, m/e (%) 412 (5.0), 373 (73,344 (12.5), 333 (15), 303 (27.5), 275 (25),
259 (20), 207 (30), 183 (70), 165 (49, 135 (15), 109 (20), 81 (52.5), 69 (100). IR
2980,2940,2150,1670,1450,1380,1600,1150 and 820. Anal. Calcd for C2gH44S
C, 81.55, H, 10.67. Found C, 81.57, H, 10.61.
Preparation of 87. To a suspension of LiAlH4 (0.6 g, 4.0 equiv.) in 10 mL
of anhyd.ether was added dropwise 86 (6.62 g, 16.0 mmol) in 10 mL of ether over a
period of 30 mi . . The reaction was refluxed for 18 hr then cooled to O•‹C. Excess hydride
was destroyed by careful addition of 1 N HCL The precipitated salts were removed by
filtration through 2 cm of Celite. The filtrate was washed with saturated NaCl solution
and dried over anhyd.MgSO4. Removal of the solvent in vacuo followed by flash
column chromatography using hexaneICH2Cl~ (15/1) as eluant gave 6.0 g (90%) of 87.
IH NMR (CDCl3) 6 1.6 (s, 6H. vinyl methyl), 1.62 (s, 9H, vinyl methyl), 1.67 (s, 6H,
vinyl methyl), 1.98-2.10 (my 12H, Cy, Cg: Cg, C7, Cloy C1 I-CH~), 2.4 (dt, 2H,
Cz9-CH2 J = 7 Hz), 2.72 (t, 2H, C1'-CH2 J = 7 Hz), 3.10 (ddd, 2H, C3-CH2 J = 11
Hz, 6 Hz, 1.8 Hz), 5.1-5.2 (my 5H, vinyl H), 5.8 (q, lHy C2-vinyl H J = 18 Hz, 7
Hz), 6.1 (d, 1Hy C1-vinyl H J = 18 Hz, 1.9 Hz). MS, m/e (%) 414 (57.5)- 396 (27.5),
329 (43, 3 14 (52.5), 239 (30), 2 1 1 (50), 109 (27.5), 8 1 (25)- 68 (100). Anal. Calcd
for C2gH46S C, 81.15, H, 11.11. Found C, 81.17, H, 11.21.
Synthesis of 9. The sulfide 87 (0.6 g, 1.6 mmol) was added to 2 mL of dry
methyl iodide in a schlenk tube, and the reaction mixture was allowed to stand at rt far
overnight. It was then heated with ether, methanol and ethyl acetate to remove excess
methyl iodide and dissolve the oily residue. Crystallization occured upon cooling to yield
0.16 g (27%) of the desired product. IH NMR (CDC13) 6 1.6 (s, 6H, vinyl methyl),
1.62 (s, 9H, vinyl methyl), 1.67 (s, 6H, vinyl methyl), 1.9 (s, 3H, S-methyl), 2.20 (m,
12Hy Cg', Cg', Cg, C7y C10, Cll-CH2)y 2.4 (dt, 2H, C2'-CH2 J = 7 HZ), 2.72 (t,
2H, C1'-CH2 J = 7 Hz), 3.15 (ddd, 2H, C3-CH2 J = 11 HZ, 6 HZ, 1.8 Hz), 5.1-5.2
(m, 5H, vinyl H), 5.8 (q, lH, C2-vinyl H J = 18 Hz, 7 Hz), 6.1 (d, 1 H, C1 -vinyl H J
= 18 Hz, 1.9 Hz). MS, FAB (Xenon/sulfolane) 429 (M+, 100).
REFERENCES
(1) (a) Wolfenden, R.Transition States of Biochemical Processes Plenum Publishing
Co., New York, 1978. (b) Engel, P.A. Enzyme Kinetics: The Steady-State Approach
Chapman Hall & Methuen Inc., New York, 1981 .
(2)The Biochemistry of Artherosclerosis, Vol. 7 The Biochemistry of Disease Marcel
Dekker Inc., New York, 1979 .
(3) Brown, M.S.; Goldstein, J.L.Scientific American 1984, 52.
(4) Poulter, C.D.; Rilling, H.C. Biosynthesis of Isoprenoid Compounds John Wiley
and Sons., New York, 1981, 1 .
(5) (a) Avruch, L.; Fischer, S.; Pierce, H.D.Jr.; Oehlschlager, A.C.Can. J. Biochem.
1976,54, 657. (b) Oehlschlager, A.C.; Pierce, H.D.Jr.; Pierce, A.M.; Angus, R.H.;
Quantin-Martenot, E.; Unrau, E.M.; Srinivasan, R. Biogenesis and Function of Plant
Lipids Elsevier/North-Holland Biomedical Press, Amsterdam, 1980,395. (c)
Oehlschlager, A.C.; Angus, R.H.; Pierce, A.M.; Pierce, H.D.Jr.; Srinivasan, R.
Biochem. 1984,23, 3582. (d) Acuna-Johnson, P.; Czyzewska, E.; Oehlschlager,
A.C.; Pierce, H.D.Jr.; Pierce, A.M. unpublished results .
(6) (a) Rahier, A.; Taton, M.; Schmitt, P.; Benveniste, P.; Place, P.; Anding, C.
Phytochernistry, 1985,24, 1223. (b) Rahier, A.; Bouvier, P.; Cattel, L.; Narula, A.;
Benveniste, P. Biochem. Soc. Trans. 1983,11, 537. (c) Delprino, L.; Balliano, G.;
Cattel, L.; Benveniste P.; Bouvier, P. J. Chem. Soc., Chem. Commun., 1981, 381.
(7) (a) Narula, A.S.; Rahier, A,; Benveniste, P.; Place, P.; Anding, C. J. Am. Chem.
Soc, 1981,103, 2408. (b) Rahier, A,, Genot, J.C.; Schuber, F.; Benveniste P.;
Narula, A.S. J. Biol. Chem. 1984,259, 15215. (c) Cerutti, M.; Delprino, L.; Cattel,
L.; Bouvier-Nave, P.; Duriatti, A.; Schuber, F.; Benveniste, P. J . Chem. SOC., Chem.
Commun., 1985, 1054.
(8) Reardon, J.E.; Abeles, R.H. Biochem. 1985,25, 5609.
(9) Croteau, R.; Wheeler, C.J.; Aksela, R.; Oehlschlager, A.C. J. Biol. Chem. 1986,
261,7257.
(10) (a) Rilling, H.C.; Poulter, C.D.; Epstein, W.W.; Larsen, B. J. Am. Chem. Soc.
1971,93, 1783. (b) Ortiz de Montellano, P.R.; Wei, J.S.; Castillo, R.; H s ~ , C.K.;
Boparai, A. J. Medicinal Chem. 1977,20,243. (c) van Tamelen, E.E.; Schwartz,
M.A. J. Am. Chem. Soc. 1971,93, 1780. (d) Popjak, G.; Agnew, W.S. Mol. Cell
Biochem. 1979,27, 97. (e) Agnew, W.S.; Popjak, G. J. Biol. Chem. 1978, 253,
4566. (f) Kuswik-Rabiega, K.; Rilling, H.C.; J. Biol. Chem. 1987,262, 1505. (g)
Cornforth, J.W. Chem. Rev. 1973,2, 1. (h) Cornforth, J.W. Angew. Chem. Int. Ed.
Engl. 1968, 7, 903. (i) Bertolino, L.; Altman, L.J.; Vasak, J.; Rilling, H.C.;
Biochim. Biophys. Acta 1978,530, 17. (i) Altman, L.J.; Kowerski, R.C.; Rilling,
H.C. J. Am. Chem. Soc. 1971, 93, 1782. (k) Agnew, W.S.; Popjak, G. J. Biol
Chem. 1978,253, 4566. (1) Altman, L.J.; Kowerski, R.C.; Laungani, D.R. J. Am.
Z k m . Soc. 1978,100, 6174. (m) Agnew, W.S.; Popjak, G. J. Biol Chem. 1978,
(1 1) (a) Saxidifer, R.M.; Thompson, M.D.; Gaughan, R.G.; Poulter, C.D. J. Am.
Chem. Soc. 1982, I04, 7376. (b) Capson, T.L.; Thompson, M.D.; Dixit, V.M.;
Gaughan, R.G.; Poulter, C.D. J. Org. Chem. 1988,53, 5903.
(12) Poulter, C.D.; Capson, T.L.; Thompson, M.D.; Bard, R.S. J. Am. Chem. Soc.
1989,111, 3734. The inhibition studies will be executed by Mrs. Armin Samiei at
SFU.
(13) Leopold, E.J. Org Syn. Prep. available on request.
(14) (a) Mancuso, A.J.; Huang, S.L.; Swern, D. J. Org. Chem. 1978,43, 2480. (b)
Other methods of preparing pure tram-geranial including Collins oxidationc and use of
activated Mn@ require large excess of reagents.d (c) Ratcliffe, R.W.Organic Syn.
1976,55, 84. (d) Corey, E.J.; Gilrnan, N.W.; Ganem, B.E.; J. Am. Chem. Soc.
1968,90, 5616.
(15) (a) Attempts to use the anion f-ed from DMSO and ~ a ~ b or the use of KO'Bu in
DMSOc has been reported to result in 10-20% of Z isomer of triene. Use of butyl
lithium as base gave only 45% yield. (b) Greenwald, R.; Chaykovsky, M.; Corey, E.J.
J. Org. Chem. 1963,28, 1128. (c) Denney, D.B.; Song, J. J. Org. Chem. 1964,29,
495.
(16) (a) Brown, H.C.; Zweifel, G. J. Am. Chem. Soc. 1961,83, 1241. (b) Brown,
H.C. Organic Synthesis via Boranes, Wiley Interscience, New York, 1975, p 29, 35,
85 and 241.
(17) (a) Davis, R.; Untch, K.G J. Org. Chem. 1981, 46,2985. (b) Mizuno, A.;
Hamada, A.Y .; Shioiri, T. Syn. Commun. 1980, 1007. (c) Nystrom, R.F.; Brown,
W.G. J . Am. Chem. Soc. 1969,69, 2548.
(18) (a) Osbond, J.M. J. Chem. Soc. 1961, 5270. (b) Katzenellenbogen, J.A.;
Lenox, R.S. J. Org. Chem. 1973, 38, 326. (c) Coates, R.M.; Ley, D.A.; Cavender,
D.L. J. Org. Chem. 1978,43, 4915.
(19) Hoye, T.R.; Kurth, M.J. J. Org. Chem. 1978,43, 3693.
(20) Kojima, Y.; Wakita, S.; Kato, N.Tetrahedron Lett. 1979, 4577.
(21) Mitsunobu, 0.; Eguchi, M. Bull. Chem. Soc. Japan 1971,44, 3427.
(22) Volante, R.P.Tetrahedron Lett. 1981, 31 19.
(23) Bobbio, P.A. J. Org. Chem. 1961, 26, 3023.
(24) (a) Weiler, L. J. Am. Chem. Soc. 1970, 92, 6702. (b) Sum, F.W.; Weiler, L.
Can. J. Chem. 1979,57, 1431. (c) Weiler, L.; Spino, C.; Alderdice, M. Tetrahedron
Lett. 1984, 1643.
(25) Brown, H.C.; Krishnamurthy, S. Tetrahedron 1979,35, 567.
(26) (a) Masamune, S.; Kamata, S.; Schilling, W. J. Am. Chem. Soc. 1975, 97,
35 15. (b) Detty, M.R.; Wood, G.P. J. Org. Chem. 1980,45, 80.
(27) Reich, H.J.; Cohen, M.L. J. Am. Chem. Soc. 1979,101, 1307.
(28) Arnold, R.T.; Kulenovic, S.T. J. Org. Chem. 1978,43, 3687. (b) Kotake, H.;
Inomato, K.; Aoyamo, S.I.; Kinoshita, H. Chem. Lett. 1977,73. (c) Garst, M.E.;
Mcbride, B.J.; Johnson, A.T. J. Org. Chem. 1983,48, 8.
(29) Hooz, J.; Gilani, S.S.H.Can. J. Chem. 1968,46, 86.
(30) (a) Scholz, D. Liebigs Ann. Chem., 1984, 259. (b) Hayashi, S.; Furukawa, M,;
Yamarnoto, Y.; Niigata, K. Chem. Pharm. Bull. 1967,15, 1 188. (c) Wiley, G.A.;
Hershkowitz, R.L.; Rein, B.M.; Chung, B.C. J. Am. Chem. Soc. 1964, 86, 964.
(3 1) Kuwajima, I.; Doi, Y. Tetrahedron Lett., 1972, 1 163.
(32) (a) Millar, J.G.; Oehlschlager, A.C.; Wong, J.W. J. Org. Chem. 1983,48, 4404.
(b) Posner, G.H.; Ting, J-S.; Lentz, C.M.Tetrahedron 1976,32,228 1.
(33) Coates, R.M.; Robinson, W.H. J. Am. Chem. Soc. 1971,93, 1785.
(34) Corey, E.J.; Wollenberg, R.H. J. Org. Chem. 1975,40, 2265.
(35) Greene, T.W. Protective Groups in Organic Synthesis Wiley Interscience, New
York, 1980.
(36) Joshi, G.C.; Singh, N.C.; Pande, L.M.Tetrahedron Lett. 1971, 1461.
(37) Jomtsma, R.; Steinberg, H.; de Boer, T.J. Recl. Trav. Chim. Pays-Bas 1981,
100, 184.
(38) Westmijze, H.; Ruitenberg, K.; Meijer, J.; Vermeer, P. Tetrahedron Lett., 1983,
2797.
(39) Piers, E.; Jean, M.; Mans, P.S. Tetrahedron Lett. 1987, 5075.
(40) Lipshutz, B.H.; Parker, D.; Kozlowski, J.A. J. Org. Chem. 1983,48, 3334.
(41) van Horn, D.E.; Negishi, Ei-ichi J. Am. Chem. Soc. 1978, 100, 2252.
(42) (a) Brandsma, L. Preparative Acetylenic Chemistry. Elsevier Publishing Co., New
York, 1971. (b) Brandsma, L.; Verkruijsse, H.D.Synthesis of Acetylenes, Allenes and
Cummulenes Elsevier Publishing Co., New York 1981 .
(43) Venneer, P.; Meijer, J.; Eylander, C.; Brandsma, L. Recl. Trav. Chim. Pays-Bas.
1976,95, 26.
(44) (a) Baba, S.; van Horn, D.E.; Negishi, E.I.Tetrahedron, 1976, 1927. (b)
Kozikowski, A.P.; Ames, A.; Wetter, H. J. Organornet. Chem. 1979, I69, C33. (c)
Tiecco, M.; Testaferri, L.; Tingolo, M.; Chianelli, D.; Monlanacci, M. J. Org. C k z .
1983,48, 4795. (d) Murahashi, S.I.; Yamamura, M.; Yanagisawa, K.; Mita, N.;
Kondo, K. J. Org. Chem. 1979,44, 2408. (e) Posner, G.H.; Tang, P.W. J. Org.
Chem. 1978,43, 4131. (f) Alexakis, A.; Normant, J.F. Synthesis 1985, 72.
(45) (a) Corey, E.J.; Cashman, J.R.; Meikrich, T.; Corey, D.R. J. Am. Chem. Soc.
1985,107, 7 13. (b) Guss, C.O.; Chamberlain, D.L. J. Am. Chem. Soc. 1952,74,
1342.
(46) Miyaura, N.; Yarnada, K.; Suginoine, H.; Suzukui, A. J. Am. Chem. Soc. 1985,
107, 972.
(47) Still, W.C; Kahn, M.; Mitra, A. J. Org. Chem. 1978,43, 2923.